From a39841133e15cddd2c3f1c0e2731c60502b41c2a Mon Sep 17 00:00:00 2001 From: June Choe Date: Wed, 6 Sep 2023 10:57:40 -0400 Subject: [PATCH] render site --- docs/articles/Garrison-et-al-2020.html | 4 ++-- docs/articles/Geller-et-al-2020.html | 6 +++--- docs/articles/asynchronous-cpa.html | 18 +++++++++++------- docs/articles/tidying-output.html | 2 +- docs/news/index.html | 5 +++-- docs/pkgdown.yml | 2 +- docs/search.json | 2 +- 7 files changed, 22 insertions(+), 17 deletions(-) diff --git a/docs/articles/Garrison-et-al-2020.html b/docs/articles/Garrison-et-al-2020.html index 14d4293..02abb1f 100644 --- a/docs/articles/Garrison-et-al-2020.html +++ b/docs/articles/Garrison-et-al-2020.html @@ -887,7 +887,7 @@

Repeating the CPA with logis ) }) #> user system elapsed -#> 0.00 0.07 5.63 +#> 0.27 0.58 17.53 CPA_agg_1000 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── <null_cluster_dists> ── @@ -911,7 +911,7 @@

Repeating the CPA with logis ) }) #> user system elapsed -#> 0.12 0.41 12.87 +#> 0.14 0.20 31.39 CPA_trial_1000 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2.8) ──────────── <null_cluster_dists> ── diff --git a/docs/articles/Geller-et-al-2020.html b/docs/articles/Geller-et-al-2020.html index cac8a4a..3a50a9e 100644 --- a/docs/articles/Geller-et-al-2020.html +++ b/docs/articles/Geller-et-al-2020.html @@ -479,7 +479,7 @@

D) Specifying random effects) }) #> user system elapsed -#> 0.03 0.06 13.65 +#> 0.06 0.04 12.13 re_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── <null_cluster_dists> ── @@ -558,7 +558,7 @@

D) Specifying random effects) }) #> user system elapsed -#> 0.06 0.08 12.04 +#> 0.05 0.03 12.99 re_max_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── <null_cluster_dists> ── @@ -714,7 +714,7 @@

E) Contrast coding) }) #> user system elapsed -#> 0.05 0.08 5.67 +#> 0.03 0.11 5.28 final_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── <null_cluster_dists> ── diff --git a/docs/articles/asynchronous-cpa.html b/docs/articles/asynchronous-cpa.html index b34f19f..fa03571 100644 --- a/docs/articles/asynchronous-cpa.html +++ b/docs/articles/asynchronous-cpa.html @@ -244,8 +244,12 @@

Creating the future object#> Elapsed: 1.0s #> Elapsed: 1.5s #> Elapsed: 2.0s -#> Elapsed: 2.5s -

We reach the end of the while loop at 2.5 seconds, which is +#> Elapsed: 2.5s +#> Elapsed: 3.0s +#> Elapsed: 3.5s +#> Elapsed: 4.0s +#> Elapsed: 4.5s +

We reach the end of the while loop at 4.5 seconds, which is approximately how long the CPA took to complete (we confirm this in the next section). At this point the future has completed and its result is available for collection.

@@ -275,22 +279,22 @@

Collecting the results 
 CPA_async <- value(f)
-#> Connecting to Julia TCP server at localhost:11980 ...
+#> Connecting to Julia TCP server at localhost:11981 ...
 #> ℹ Detecting empirical clusters and calculating cluster-mass statistics.
-#> ✔ Detecting empirical clusters and calculating cluster-mass statistics. [75ms]
+#> ✔ Detecting empirical clusters and calculating cluster-mass statistics. [125ms]
 #> 
 #> ℹ Sampling cluster-mass statistics from a bootstrapped null distribution.
-#> ✔ Sampling cluster-mass statistics from a bootstrapped null distribution. [2s]
+#> ✔ Sampling cluster-mass statistics from a bootstrapped null distribution. [4.8s]
 #> 
 #> ℹ Calculating the probability of the observed cluster-mass statistics.
-#> ✔ Calculating the probability of the observed cluster-mass statistics. [24ms]
+#> ✔ Calculating the probability of the observed cluster-mass statistics. [28ms]
 #>

Note how value() also prints the messages encountered while executing the CPA code. We see this because we did not specify progress = FALSE in the clusterpermute() call passed to the future. For our purposes, just note that adding up the times from the messages is similar to what we saw from the while loop -(2.5 seconds).

+(4.5 seconds).

The output of the asynchronous CPA (CPA_async) is identical to the initial CPA that we ran at the beginning (CPA) because both were ran with the default seed of 1 and diff --git a/docs/articles/tidying-output.html b/docs/articles/tidying-output.html index 0ef8c82..526bb7d 100644 --- a/docs/articles/tidying-output.html +++ b/docs/articles/tidying-output.html @@ -372,7 +372,7 @@

In other contextsfull_test <- clusterpermute(jlmer_spec, threshold = 1, nsim = 1000, progress = FALSE) }) #> user system elapsed -#> 0.02 0.05 2.02 +#> 0.00 0.01 1.77 full_test #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── <null_cluster_dists> ── diff --git a/docs/news/index.html b/docs/news/index.html index 43c5850..f4f3dbf 100644 --- a/docs/news/index.html +++ b/docs/news/index.html @@ -66,9 +66,10 @@

Case Studies

jlmerclusterperm (development version)

-

jlmerclusterperm 1.0.4

+

jlmerclusterperm 1.0.4

CRAN release: 2023-09-06

No user-visible changes

-
  • Restructured internal Julia code into a module (JlmerClusterPerm.jl)
    • +

    • Restructured internal Julia code into a module (JlmerClusterPerm.jl)

      • +

    • Upon activating project, Manifest.toml is cached to tools::R_user_dir() to speed up pre-compilation.

jlmerclusterperm 1.0.3

CRAN release: 2023-07-19

diff --git a/docs/pkgdown.yml b/docs/pkgdown.yml index 0cb8e98..55e994c 100644 --- a/docs/pkgdown.yml +++ b/docs/pkgdown.yml @@ -10,7 +10,7 @@ articles: reproducibility: reproducibility.html tidying-output: tidying-output.html jlmerclusterperm: jlmerclusterperm.html -last_built: 2023-09-01T17:28Z +last_built: 2023-09-06T12:45Z urls: reference: https://yjunechoe.github.io/jlmerclusterperm/reference article: https://yjunechoe.github.io/jlmerclusterperm/articles diff --git a/docs/search.json b/docs/search.json index bf6bbb7..e6e6680 100644 --- a/docs/search.json +++ b/docs/search.json @@ -1 +1 @@ -[{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Asynchronous CPA","text":"Minimal example data CPA specification: Example CPA (1000 simulations): details actual CPA skipped ’s focus vignette, example readme just nsim.","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) chickweights <- ChickWeight chickweights$Time <- as.integer(factor(chickweights$Time)) chickweights_spec <- make_jlmer_spec( formula = weight ~ 1 + Diet, data = chickweights, subject = \"Chick\", time = \"Time\" ) set_rng_state(123L) CPA <- clusterpermute( chickweights_spec, threshold = 2.5, nsim = 1000, progress = FALSE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"basic-idea","dir":"Articles","previous_headings":"","what":"Basic idea","title":"Asynchronous CPA","text":"basic idea behind asynchronous CPA strategy start background R process whose sole job send instructions CPA Julia. ’s asynchronous running CPA way block interactive R session. Since work done Julia anyways, parallelization virtually impact performance evaluating R code. future package allows asynchronous evaluation R code. ’s big package implementing complex topic. can read project futureverse just show bare minimum . advanced users, note ’re use parallelization asynchronous properties (non-blocking evaluation R code background process). recommended start multiple R processes running CPA Julia session shared already multithreaded (can spare cores, set options(\"jlmerclusterperm.nthreads\") calling jlmerclusterperm_setup()). Later updates jlmerclusterperm may wrap workflow principled way, now vignette serves minimally working example asynchronously running CPA.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"setup-for-async-cpa","dir":"Articles","previous_headings":"Walkthrough","what":"Setup for async CPA","title":"Asynchronous CPA","text":"Three things order workflow: Load future package Initialize multisession future Grab options jlmerclusterperm package environment Please treat .jlmerclusterperm internal variable read-object - ’s unexported meant manipulated.","code":"library(future) plan(multisession) pkgopts <- as.list(jlmerclusterperm:::.jlmerclusterperm)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"creating-the-future-object","dir":"Articles","previous_headings":"Walkthrough","what":"Creating the future object","title":"Asynchronous CPA","text":"start creating special object class using future::future(). can simply pass clusterpermute() code first argument future(), slight modification ensure background process connects Julia session. template follows. First, make pkgopts object (defined ) available future (via globals argument). , inside future expression ensure background process shares Julia session (via list2env(...)). future object replicating 1000-simulation CPA look like following: future created, immediately starts executing code background process. important evaluate future object (f) directly. Instead, query future::resolved() - simply tells whether background evaluation completed : , use loop show background CPA non-blocking. evaluating R code interactive session simultaneously CPA running: reach end loop 2.5 seconds, approximately long CPA took complete (confirm next section). point future completed result available collection. Although R code just crude progress alert using Sys.sleep(), can freely evaluate R code except functions call Julia. can make plots, clean data, write analyses, etc., just don’t run another CPA one already running Julia process shared.","code":"# Not run future( { list2env(pkgopts, jlmerclusterperm:::.jlmerclusterperm) ## Your CPA code below ## }, globals = structure(TRUE, add = \"pkgopts\") ) f <- future( { list2env(as.list(pkgopts), jlmerclusterperm:::.jlmerclusterperm) ## Your CPA code below ## set_rng_state(123L) clusterpermute( chickweights_spec, threshold = 2.5, nsim = 1000 ) }, globals = structure(TRUE, add = \"pkgopts\") ) resolved(f) #> [1] FALSE i <- 0 while (!resolved(f)) { Sys.sleep(0.5) i <<- i + 1 cat(sprintf(\"Elapsed: %.01fs\", i / 2), \"\\n\") } #> Elapsed: 0.5s #> Elapsed: 1.0s #> Elapsed: 1.5s #> Elapsed: 2.0s #> Elapsed: 2.5s resolved(f) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"collecting-the-results","dir":"Articles","previous_headings":"Walkthrough","what":"Collecting the results","title":"Asynchronous CPA","text":"can collect output background process future::value(). first, double check make sure background process indeed finished evaluating: ’s confirmed, can use value() collect results future assign variable inspection: Note value() also prints messages encountered executing CPA code. see specify progress = FALSE clusterpermute() call passed future. purposes, just note adding times messages similar saw loop (2.5 seconds). output asynchronous CPA (CPA_async) identical initial CPA ran beginning (CPA) ran default seed 1 RNG counter value 123L: separate vignette covers Julia RNG. practice, may want use random seed (via set_rng_seed()) background CPA.","code":"resolved(f) #> [1] TRUE CPA_async <- value(f) #> Connecting to Julia TCP server at localhost:11980 ... #> ℹ Detecting empirical clusters and calculating cluster-mass statistics. #> ✔ Detecting empirical clusters and calculating cluster-mass statistics. [75ms] #> #> ℹ Sampling cluster-mass statistics from a bootstrapped null distribution. #> ✔ Sampling cluster-mass statistics from a bootstrapped null distribution. [2s] #> #> ℹ Calculating the probability of the observed cluster-mass statistics. #> ✔ Calculating the probability of the observed cluster-mass statistics. [24ms] #> identical(CPA_async, CPA) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"background","dir":"Articles","previous_headings":"","what":"Background","title":"de Carvalho et al. 2021","text":"data comes eye-tracking study de Carvalho, Dautriche, Fiévet, & Christophe (2021) “Toddlers exploit referential syntactic cues flexibly adapt interpretation novel verb meanings.” article reproduces expands Experiment 2 analysis, used CPA (t-tests) conduct pairwise comparison three conditions (relevant Figure 7 paper ). data comes mostly ready CPA straight study’s OSF repository. collapsible chunk contains code prepare data reproduce figure. Data cleaning Code Figure 7 plots data original study (close reproduce ). E2_data_agg dataframe following four columns: Subject: Unique identifier subjects Condition: -participant factor variable three levels (\"Intransitive\", \"RightDislocated\", \"Transitive\") Time: continuous measure time 0-8000ms 50ms intervals Prop: Proportion looks target (averaged across trials within condition, subject) following reproduction Figure 7 original paper: original study used pairwise comparisons analyze relationship three conditions. smaller figures (B, C, D) plot following relationships: (B) looks target \"Transitive\" compared \"RightDislocated\" (C) looks target \"RightDislocated\" compared \"Intransitive\" (D) looks target \"Transitive\" compared \"Intransitive\" rest vignette organized follows: First, replicate individual pairwise CPAs original paper. Next, consider parsimonious analysis reformulates research hypothesis choice regression contrast. Lastly, build model (2) consider issues around model diagnostics complexity (linear vs. logistic; fixed vs. mixed) context CPA Load package start Julia instance jlmerclusterperm_setup() proceeding.","code":"library(dplyr) library(forcats) filepath <- \"https://osf.io/download/var2h/?view_only=2a06746ba360446e9857df73307d21be\" E2_data_raw <- readr::read_delim(filepath) E2_data <- E2_data_raw %>% filter(away != 1) %>% mutate(Target = as.integer(bin_2P == 1)) %>% mutate(Condition = fct_recode( factor(Condition, levels = c(\"intrans\", \"pros\", \"trans\")), \"Intransitive\" = \"intrans\", \"RightDislocated\" = \"pros\", \"Transitive\" = \"trans\" )) %>% select(Subject, Trial, Condition, Time, Target) E2_data_agg <- E2_data %>% group_by(Subject, Condition, Time) %>% summarize(Prop = mean(Target), .groups = \"drop\") # Unaggregated trial-level data of 1s and 0s E2_data #> # A tibble: 69,123 × 5 #> Subject Trial Condition Time Target #> #> 1 200.asc 0 RightDislocated 0 0 #> 2 200.asc 0 RightDislocated 400 0 #> 3 200.asc 0 RightDislocated 450 1 #> 4 200.asc 0 RightDislocated 500 1 #> 5 200.asc 0 RightDislocated 550 1 #> 6 200.asc 0 RightDislocated 600 1 #> 7 200.asc 0 RightDislocated 650 1 #> 8 200.asc 0 RightDislocated 700 1 #> 9 200.asc 0 RightDislocated 750 1 #> 10 200.asc 0 RightDislocated 800 1 #> # ℹ 69,113 more rows # Aggregated subject-mean proportions data used in original study E2_data_agg #> # A tibble: 11,540 × 4 #> Subject Condition Time Prop #> #> 1 200.asc RightDislocated 0 0.5 #> 2 200.asc RightDislocated 50 1 #> 3 200.asc RightDislocated 100 1 #> 4 200.asc RightDislocated 150 1 #> 5 200.asc RightDislocated 200 1 #> 6 200.asc RightDislocated 250 0.5 #> 7 200.asc RightDislocated 300 0.5 #> 8 200.asc RightDislocated 350 0.4 #> 9 200.asc RightDislocated 400 0.286 #> 10 200.asc RightDislocated 450 0.429 #> # ℹ 11,530 more rows library(ggplot2) #> Warning: package 'ggplot2' was built under R version 4.3.1 make_fig7_plot <- function(conditions) { E2_data %>% group_by(Condition, Time) %>% summarize( Prop = mean(Target), se = sqrt(var(Target) / n()), lower = Prop - se, upper = Prop + se, .groups = \"drop\" ) %>% filter(Condition %in% conditions) %>% ggplot(aes(Time, Prop, color = Condition, fill = Condition)) + geom_hline(aes(yintercept = .5)) + geom_ribbon( aes(ymin = lower, ymax = upper), alpha = .2, show.legend = FALSE, ) + geom_line(linewidth = 1) + scale_color_manual( aesthetics = c(\"color\", \"fill\"), values = setNames(scales::hue_pal()(3)[c(2, 1, 3)], levels(E2_data$Condition)) ) + scale_y_continuous(limits = c(.2, .8), oob = scales::oob_keep) + labs(y = NULL, x = NULL) + theme_minimal() + theme(axis.title.y = element_text(angle = 0, vjust = .5, hjust = 0)) } fig7_comparisons <- list( \"A\" = c(\"Intransitive\", \"RightDislocated\", \"Transitive\"), \"B\" = c(\"RightDislocated\", \"Transitive\"), \"C\" = c(\"Intransitive\", \"RightDislocated\"), \"D\" = c(\"Intransitive\", \"Transitive\") ) fig7 <- lapply(fig7_comparisons, make_fig7_plot) # Figure 7 combined plots library(patchwork) #> Warning: package 'patchwork' was built under R version 4.3.1 p_top <- fig7$A + scale_x_continuous(breaks = scales::breaks_width(1000)) + theme(legend.position = \"bottom\") p_bot <- (fig7$B + fig7$C + fig7$D) & guides(color = guide_none()) fig7_combined <- p_top / guide_area() / p_bot + plot_layout(guides = \"collect\", heights = c(2, .1, 1)) + plot_annotation(tag_levels = \"A\") E2_data_agg #> # A tibble: 11,540 × 4 #> Subject Condition Time Prop #> #> 1 200.asc RightDislocated 0 0.5 #> 2 200.asc RightDislocated 50 1 #> 3 200.asc RightDislocated 100 1 #> 4 200.asc RightDislocated 150 1 #> 5 200.asc RightDislocated 200 1 #> 6 200.asc RightDislocated 250 0.5 #> 7 200.asc RightDislocated 300 0.5 #> 8 200.asc RightDislocated 350 0.4 #> 9 200.asc RightDislocated 400 0.286 #> 10 200.asc RightDislocated 450 0.429 #> # ℹ 11,530 more rows fig7_combined library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"replicating-the-pairwise-cpas","dir":"Articles","previous_headings":"","what":"Replicating the pairwise CPAs","title":"de Carvalho et al. 2021","text":"three conditions experiment levels Condition factor variable: begin replication \"Transitive\" vs. \"RightDislocated\" comparison shown Figure 7B apply logic two pairwise comparisons 7C 7D.","code":"levels(E2_data_agg$Condition) #> [1] \"Intransitive\" \"RightDislocated\" \"Transitive\""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"transitive-vs--rightdislocated","dir":"Articles","previous_headings":"Replicating the pairwise CPAs","what":"Transitive vs. RightDislocated","title":"de Carvalho et al. 2021","text":"reproduced Figure 7B compares \"Transitive\" \"RightDislocated\" conditions. paper (caption Figure 7; emphasis mine) reports: transitive right-dislocated conditions differed second repetition novel verbs (~6400 ms onset test trials end trials). now replicate analysis. First, prepare specification object. Two things note : express original t-test regression model Condition predictor drop third, unused condition data factor representation Next, fit global model sanity check structure model output. get one estimate ConditionTransitive positive coefficient, ’d expect: Finally, call clusterpermute() threshold = 1.5 (original study) simulate 100 permutations: detect largest empirical cluster spanning 6400-8000ms reported original paper. converges around p=0.02 separate 10,000-simulation run (shown ).","code":"fig7$B spec_7B <- make_jlmer_spec( formula = Prop ~ Condition, data = E2_data_agg %>% filter(Condition %in% c(\"Transitive\", \"RightDislocated\")) %>% mutate(Condition = droplevels(Condition)), # or forcats::fct_drop() subject = \"Subject\", time = \"Time\" ) spec_7B #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: Prop ~ 1 + ConditionTransitive #> Predictors: #> Condition: ConditionTransitive #> Groupings: #> Subject: Subject #> Trial: #> Time: Time #> Data: #> # A tibble: 7,688 × 4 #> Prop ConditionTransitive Subject Time #> #> 1 0.5 0 200.asc 0 #> 2 1 0 200.asc 50 #> 3 1 0 200.asc 100 #> # ℹ 7,685 more rows #> ──────────────────────────────────────────────────────────────────────────────── jlmer(spec_7B) #> #> ────────────────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ────────────────────────────────────────────────────────────────────────────────── #> (Intercept) 0.592999 0.00364424 162.72 <1e-99 0.585856 0.600141 #> ConditionTransitive 0.0569017 0.0051591 11.03 <1e-27 0.04679 0.0670133 #> ────────────────────────────────────────────────────────────────────────────────── clusterpermute(spec_7B, threshold = 1.5, nsim = 100L, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionTransitive (n = 100) #> Mean (SD): -2.674 (27.83) #> Coverage intervals: 95% [-61.740, 46.439] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionTransitive #> [300, 1150]: 40.643 (p=0.1287) #> [2700, 2750]: -4.068 (p=0.9010) #> [6150, 6200]: 3.249 (p=0.9505) #> [6400, 8000]: 88.742 (p=0.0198) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"rightdislocated-vs--intransitive","dir":"Articles","previous_headings":"Replicating the pairwise CPAs","what":"RightDislocated vs. Intransitive","title":"de Carvalho et al. 2021","text":"reproduced Figure 7C compares \"RightDislocated\" \"Intransitive\" conditions. paper reports: intransitive right-dislocated conditions differed first repetition novel verbs (2100 ms 3500 ms beginning test trials). repeat CPA procedure pairwise comparison: largest empirical cluster detect spans 2150-3650ms, slightly different cluster reported original paper (2100-3500ms). relatively less “extreme” cluster converges around p=0.05 10,000-simulation run.","code":"fig7$C spec_7C <- make_jlmer_spec( formula = Prop ~ Condition, data = E2_data_agg %>% filter(Condition %in% c(\"RightDislocated\", \"Intransitive\")) %>% mutate(Condition = droplevels(Condition)), subject = \"Subject\", time = \"Time\" ) clusterpermute(spec_7C, threshold = 1.5, nsim = 100L, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionRightDislocated (n = 100) #> Mean (SD): 0.619 (33.26) #> Coverage intervals: 95% [-63.734, 67.326] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionRightDislocated #> [150, 200]: 3.562 (p=0.9010) #> [2150, 3650]: 72.842 (p=0.0297) #> [4550, 5050]: 23.836 (p=0.3960) #> [5150, 5350]: 8.589 (p=0.7723) #> [5700, 5750]: 3.332 (p=0.9208) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"transitive-vs--intransitive","dir":"Articles","previous_headings":"Replicating the pairwise CPAs","what":"Transitive vs. Intransitive","title":"de Carvalho et al. 2021","text":"reproduced Figure 7D compares \"Transitive\" \"Intransitive\" conditions. paper reports: transitive intransitive conditions differed slightly offset first sentence test trials (4500 ms beginning test trials end trials). repeat CPA procedure pairwise comparison: largest empirical cluster detect spans 4600-8000ms, slightly different cluster reported original paper (4500-8000ms). converges around p=0.001 separate 10,000-simulation run.","code":"fig7$D spec_7D <- make_jlmer_spec( formula = Prop ~ Condition, data = E2_data_agg %>% filter(Condition %in% c(\"Transitive\", \"Intransitive\")) %>% mutate(Condition = droplevels(Condition)), subject = \"Subject\", time = \"Time\" ) clusterpermute(spec_7D, threshold = 1.5, nsim = 100L, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionTransitive (n = 100) #> Mean (SD): 2.024 (41.73) #> Coverage intervals: 95% [-70.510, 88.501] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionTransitive #> [300, 1000]: 25.928 (p=0.4356) #> [1300, 1400]: 4.808 (p=0.8713) #> [2800, 2900]: 5.356 (p=0.8218) #> [3000, 4300]: 52.638 (p=0.2079) #> [4600, 8000]: 172.635 (p=0.0198) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"expressed-as-regression-contrasts","dir":"Articles","previous_headings":"","what":"Expressed as regression contrasts","title":"de Carvalho et al. 2021","text":"now consider parsimonious analysis translates research hypothesis contrast coding avoid multiple testing. Specifically, exploit fact original paper specifies hypothesis Intransitive < RightDislocated < Transitive.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"helmert-deviation-coding","dir":"Articles","previous_headings":"Expressed as regression contrasts","what":"Helmert (deviation) coding","title":"de Carvalho et al. 2021","text":"Testing ordinal relationship levels category require possible pairwise comparisons; instead, can approximated via Helmert coding (.k.. deviance coding) K levels expressed K-1 contrasts, contrast successively comparing level vs. average previous (typically lower) levels. Critically, Helmert contrasts orthogonal, can test simultaneously single model. data, test ordinal relationship Intransitive < RightDislocated < Transitive via two contrasts: \"RightDislocated\" vs. \"Intransitive\" \"Transitive\" vs. average \"RightDislocated\" \"Intransitive\" corresponding numerical coding following: practice, Helmert contrasts often standardized deviations expressed unit 1. also comparison \"RightDislocated\" vs. \"Intransitive\" expressed 1 unit RvsI comparison \"Transitive\" vs. average \"RightDislocated\" \"Intransitive\" expressed 1 unit TvsRI: contrast matrix, make new column original data called ConditionHelm copying Condition column, apply contrasts new column: Lastly, build new specification making use full data. , predict Prop ConditionHelm estimate effect contrasts single model. sanity check, fit global model - expect estimate contrast indeed find RvsI TvsRI output positive coefficients. suggests ordinal relationship three conditions hold least globally. Note coefficient RvsI contrast exactly pairwise model using spec_7C earlier, also compared \"RightDislocated\" \"Intransitive\": spec_helm two conditions RvsI coded -0.5 0.5, spec_7C’s treatment coding coded 0 1. Since express difference \"Intransitive\" \"RightDislocated\" unit +1, two coefficients equal magnitude sign.","code":"condition_helm <- contr.helmert(3) colnames(condition_helm) <- c(\"RvsI\", \"TvsRI\") condition_helm #> RvsI TvsRI #> 1 -1 -1 #> 2 1 -1 #> 3 0 2 condition_helm[, 1] <- condition_helm[, 1] / 2 condition_helm[, 2] <- condition_helm[, 2] / 3 condition_helm #> RvsI TvsRI #> 1 -0.5 -0.3333333 #> 2 0.5 -0.3333333 #> 3 0.0 0.6666667 E2_data_agg$ConditionHelm <- E2_data_agg$Condition contrasts(E2_data_agg$ConditionHelm) <- condition_helm # For pretty-printing as fractions MASS::fractions(contrasts(E2_data_agg$ConditionHelm)) #> RvsI TvsRI #> Intransitive -1/2 -1/3 #> RightDislocated 1/2 -1/3 #> Transitive 0 2/3 spec_helm <- make_jlmer_spec( formula = Prop ~ ConditionHelm, data = E2_data_agg, subject = \"Subject\", time = \"Time\" ) spec_helm #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: Prop ~ 1 + ConditionHelmRvsI + ConditionHelmTvsRI #> Predictors: #> ConditionHelm: ConditionHelmRvsI, ConditionHelmTvsRI #> Groupings: #> Subject: Subject #> Trial: #> Time: Time #> Data: #> # A tibble: 11,540 × 5 #> Prop ConditionHelmRvsI ConditionHelmTvsRI Subject Time #> #> 1 0.5 0.5 -0.333 200.asc 0 #> 2 1 0.5 -0.333 200.asc 50 #> 3 1 0.5 -0.333 200.asc 100 #> # ℹ 11,537 more rows #> ──────────────────────────────────────────────────────────────────────────────── jlmer(spec_helm) %>% tidy(effects = \"fixed\") #> # A tibble: 3 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.591 0.00215 275. 0 #> 2 ConditionHelmRvsI 0.0640 0.00526 12.2 5.67e-34 #> 3 ConditionHelmTvsRI 0.0889 0.00457 19.5 2.02e-84 jlmer(spec_7C) %>% tidy(effects = \"fixed\") #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.529 0.00379 140. 0 #> 2 ConditionRightDislocated 0.0640 0.00536 11.9 8.30e-33"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"interpreting-cpa-results","dir":"Articles","previous_headings":"Expressed as regression contrasts","what":"Interpreting CPA results","title":"de Carvalho et al. 2021","text":"proceed normal clusterpermute() using new spec_helm: ’s summary find CPA_helm: RvsI clusters similar detected previous “partial” CPA spec_7C. slight differences cluster-mass due part due Helmert-coded model simultaneously estimating TvsRI contrast. importantly, CPA_helm detects largest cluster \"RightDislocated\" \"Intransitive\" (2150-3650ms). converges around p=0.05 10,000-simulation run. TvsRI clusters new, largest cluster predictor spans 5800ms-8000ms. converges around p=0.01 separate 10,000-simulation run. cluster effectively captures region relationship Transitive > (RightDislocated & Intransitive) emerges robust. Essentially, TvsRI comparison line \"Transitive\" invisible line runs lines \"Intransitive\" \"RightDislocated\". conclude visualizing clusters two Helmert-coded terms, annotated empirical data. “invisible” line RI Helmert coding drawn dashed line.","code":"reset_rng_state() CPA_helm <- clusterpermute(spec_helm, threshold = 1.5, nsim = 100L, progress = FALSE) CPA_helm #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionHelmRvsI (n = 100) #> Mean (SD): 4.019 (28.80) #> Coverage intervals: 95% [-45.805, 67.827] #> ConditionHelmTvsRI (n = 100) #> Mean (SD): 1.213 (24.59) #> Coverage intervals: 95% [-50.218, 47.462] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionHelmRvsI #> [150, 200]: 3.654 (p=0.8911) #> [2150, 3650]: 71.141 (p=0.0297) #> [4550, 5350]: 36.043 (p=0.2079) #> [5700, 5750]: 3.328 (p=0.9010) #> ConditionHelmTvsRI #> [300, 1050]: 35.863 (p=0.1881) #> [3500, 3700]: 8.348 (p=0.6436) #> [3850, 4250]: 15.985 (p=0.4455) #> [4800, 5600]: 36.143 (p=0.1782) #> [5800, 8000]: 123.631 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── fig7A_v2 <- fig7$A + geom_line( aes(Time, Prop, linetype = \"RI\"), inherit.aes = FALSE, data = . %>% filter(Condition %in% c(\"RightDislocated\", \"Intransitive\")) %>% group_by(Time) %>% summarize(Prop = mean(Prop)), ) + scale_linetype_manual(values = \"41\", guide = guide_legend(\"ConditionHelm\")) + guides(x = guide_none(\"\")) clusters_annotation <- tidy(CPA_helm$empirical_clusters) %>% mutate(contrast = gsub(\".*([TR])vs([RI]+)\", \"\\\\1 vs. \\\\2\", predictor)) %>% ggplot(aes(y = fct_rev(contrast))) + geom_segment( aes( x = start, xend = end, yend = contrast, color = pvalue < 0.05 ), linewidth = 8 ) + scale_color_manual(values = c(\"grey70\", \"steelblue\")) + scale_y_discrete() + scale_x_continuous(n.breaks = 9, limits = range(E2_data_agg$Time)) + theme_minimal() + theme( axis.title = element_blank(), axis.text.y = element_text(face = \"bold\"), panel.border = element_rect(fill = NA), panel.grid.major.y = element_blank() ) fig7A_v2 / clusters_annotation & plot_layout(heights = c(4, 1))"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"model-complexity","dir":"Articles","previous_headings":"","what":"Model complexity","title":"de Carvalho et al. 2021","text":"wrap case study considering complex CPA uses logistic mixed effects models trial-level data fixations target (1s 0s).","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"logistic-mixed-model","dir":"Articles","previous_headings":"Model complexity","what":"Logistic mixed model","title":"de Carvalho et al. 2021","text":"un-aggregated trial-level data use section E2_data, comes initial data preparation code chunk: apply Helmert/deviation-coded contrast matrix: specification object E2_data, add trial = \"Trial\" predict Target instead Prop. also add -subject random intercept formula: , CPA family = \"binomial\": now visualize results CPA_helm_complex CPA_helm side side: results largely , except largest, significant cluster identified TvsRI extends much complex CPA simple CPA. examine difference next.","code":"E2_data #> # A tibble: 69,123 × 5 #> Subject Trial Condition Time Target #> #> 1 200.asc 0 RightDislocated 0 0 #> 2 200.asc 0 RightDislocated 400 0 #> 3 200.asc 0 RightDislocated 450 1 #> 4 200.asc 0 RightDislocated 500 1 #> 5 200.asc 0 RightDislocated 550 1 #> 6 200.asc 0 RightDislocated 600 1 #> 7 200.asc 0 RightDislocated 650 1 #> 8 200.asc 0 RightDislocated 700 1 #> 9 200.asc 0 RightDislocated 750 1 #> 10 200.asc 0 RightDislocated 800 1 #> # ℹ 69,113 more rows E2_data$ConditionHelm <- E2_data$Condition contrasts(E2_data$ConditionHelm) <- condition_helm MASS::fractions(contrasts(E2_data$ConditionHelm)) #> RvsI TvsRI #> Intransitive -1/2 -1/3 #> RightDislocated 1/2 -1/3 #> Transitive 0 2/3 spec_helm_complex <- make_jlmer_spec( formula = Target ~ ConditionHelm + (1 | Subject), data = E2_data, subject = \"Subject\", trial = \"Trial\", time = \"Time\" ) spec_helm_complex #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: Target ~ 1 + ConditionHelmRvsI + ConditionHelmTvsRI + (1 | Subject) #> Predictors: #> ConditionHelm: ConditionHelmRvsI, ConditionHelmTvsRI #> Groupings: #> Subject: Subject #> Trial: Trial #> Time: Time #> Data: #> # A tibble: 69,123 × 6 #> Target ConditionHelmRvsI ConditionHelmTvsRI Subject Trial Time #> #> 1 0 0.5 -0.333 200.asc 0 0 #> 2 0 0.5 -0.333 200.asc 0 400 #> 3 1 0.5 -0.333 200.asc 0 450 #> # ℹ 69,120 more rows #> ──────────────────────────────────────────────────────────────────────────────── reset_rng_state() CPA_helm_complex <- clusterpermute( spec_helm_complex, family = \"binomial\", threshold = 1.5, nsim = 100, progress = FALSE ) CPA_helm_complex #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionHelmRvsI (n = 100) #> Mean (SD): 1.791 (31.12) #> Coverage intervals: 95% [-60.956, 80.601] #> ConditionHelmTvsRI (n = 100) #> Mean (SD): 2.704 (28.90) #> Coverage intervals: 95% [-54.636, 62.925] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionHelmRvsI #> [2100, 3450]: 69.563 (p=0.0495) #> [4200, 5050]: 40.460 (p=0.1584) #> [5150, 5350]: 8.379 (p=0.7228) #> ConditionHelmTvsRI #> [350, 800]: 19.358 (p=0.4158) #> [900, 1050]: 7.519 (p=0.6535) #> [1250, 1350]: 4.980 (p=0.7228) #> [3100, 4450]: 56.770 (p=0.0792) #> [4650, 8000]: 181.165 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── clusters_annotation2 <- tidy(CPA_helm_complex$empirical_clusters) %>% mutate(contrast = gsub(\".*([TR])vs([RI]+)\", \"\\\\1 vs. \\\\2\", predictor)) %>% ggplot(aes(y = fct_rev(contrast))) + geom_segment( aes( x = start, xend = end, yend = contrast, color = pvalue < 0.05 ), linewidth = 8 ) + scale_color_manual(values = c(\"grey70\", \"steelblue\")) + scale_y_discrete() + scale_x_continuous(n.breaks = 9, limits = range(E2_data$Time)) + theme_minimal() + theme( legend.position = \"bottom\", axis.title = element_blank(), axis.text.y = element_text(face = \"bold\"), panel.border = element_rect(fill = NA), panel.grid.major.y = element_blank() ) clusters_annotation / clusters_annotation2 & guides(color = guide_none()) & plot_annotation(tag_levels = list(c(\"Simple\", \"Complex\")))"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"comparison-of-cpas","dir":"Articles","previous_headings":"Model complexity","what":"Comparison of CPAs","title":"de Carvalho et al. 2021","text":"Looking timewise statistics computed simple vs. complex CPA tells us . figure plots information calls compute_timewise_statistics(): Whereas largest cluster starts emerge 5800ms CPA_helm, emerges much earlier 4650ms CPA_helm_complex. zoom region around 5800ms, see timewise statistics T vs. RI CPA_helm suddenly dip 1.5 threshold 5750ms: CPA better? existence dips spikes indicate problem, ’s consistent expectation simple CPA less robust variance. can inspect time-point model 5750ms two CPAs fitting : ’s standard way comparing goodness fit linear fixed-effects model logistic mixed-effects model fitted different data. complex model outperforms simple model classic metrics inspect glance(). doesn’t come surprise, differences largely driven number observations (nobs). Determining appropriate degree model complexity CPA beyond scope vignette, pursue discussion . Instead, conclude old wisdom: chain strong weakest link.","code":"# Compute the timewise statistics from the CPA specifications empirical_statistics <- bind_rows( tidy(compute_timewise_statistics(spec_helm)), tidy(compute_timewise_statistics(spec_helm_complex, family = \"binomial\")), .id = \"spec\" ) %>% mutate( CPA = c(\"Simple\", \"Complex\")[as.integer(spec)], Contrast = gsub(\".*([TR])vs([RI]+)\", \"\\\\1 vs. \\\\2\", predictor) ) # Time series plot of the statistics, with a line for each Helmert contrasts empirical_statistics_fig <- ggplot(empirical_statistics, aes(time, statistic)) + geom_line(aes(color = Contrast, linetype = CPA), linewidth = 1, alpha = .7) + geom_hline(yintercept = c(-1.5, 1.5), linetype = 2) + theme_classic() empirical_statistics_fig empirical_statistics_fig + geom_vline(xintercept = 5750) + coord_cartesian(xlim = 5750 + c(-500, 500), ylim = c(1, 2.5)) jlmer_simple_5750 <- to_jlmer( formula = Prop ~ ConditionHelm, data = E2_data_agg %>% filter(Time == 5750) ) jlmer_complex_5750 <- to_jlmer( formula = Target ~ ConditionHelm + (1 | Subject), family = \"binomial\", data = E2_data %>% filter(Time == 5750) ) glance(jlmer_simple_5750) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 72 4 0.223 7.51 -7.02 2.08 3.42 68 glance(jlmer_complex_5750) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 448 4 NA -301. 609. 626. 601. 444"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"background","dir":"Articles","previous_headings":"","what":"Background","title":"Garrison et al. 2020","text":"data comes eye-tracking study Garrison, Baudet, Breitfield, Aberman, & Bergelson (2020) “Familiarity plays small role noun comprehension 12-18 months”, way peekbankr corpus package. original study used growth curve model analyze children’s looks object familiar (“”) vs. unfamiliar (“”) different ages (median-split “younger” “older” groups). replicating specific analysis . Instead, follow Katie Von Holzen’s tutorial used dataset conduct simpler, cluster-based permutation analysis (CPA) differences two age groups proportion looks target. tutorial used eyetrackingR package data preparation CPA, vignette use dplyr jlmerclusterperm tasks, respectively. following set-code reads data prepares analysis: data ts_data_agg ’ll use replicate tutorial just 4 columns relevant cluster-permutation analysis: subject_id: Unique identifier subjects age: -participant factor variable (\"Younger\", \"Older\") time: continuous measure time 0-3975ms 25ms intervals prop: -subject proportion looks target object following replicates figure tutorial introduces cluster-permutation analysis:","code":"library(dplyr) remotes::install_github(\"langcog/peekbankr\", quiet = TRUE) library(peekbankr) # Load data aoi_timepoints <- get_aoi_timepoints(dataset_name = \"garrison_bergelson_2020\") administrations <- get_administrations(dataset_name = \"garrison_bergelson_2020\") # Pre-processing ## Filter for age groups of interest ps_data <- aoi_timepoints %>% left_join(administrations, by = \"administration_id\") %>% filter(!between(age, 14, 16)) %>% mutate(age_binned = ifelse(age < 14, \"Younger\", \"Older\")) ## Filter for time window of interest ts_window <- ps_data |> filter(t_norm >= 0, t_norm < 4000) ## Identify trials to exclude (trackloss in >25% of samples) to_exclude <- ts_window |> group_by(subject_id, trial_id) |> summarize(prop = mean(aoi == \"missing\"), .groups = \"drop\") |> filter(prop > 0.25) ## Exclude disqualifying trials and keep information relevant for CPA ts_data <- ts_window |> anti_join(to_exclude, by = c(\"subject_id\", \"trial_id\")) |> mutate( target = aoi == \"target\", missing = aoi == \"missing\", age = factor(age_binned, c(\"Younger\", \"Older\")) ) |> select(subject_id, age, trial_id, time = t_norm, target, missing) # Data of subject mean proportions of fixations to target ts_data_agg <- ts_data |> group_by(subject_id, age, time) |> summarize(prop = sum(target) / sum(!missing), .groups = \"drop\") # Data of trial-level fixations to target (0s and 1s) ts_data_trials <- ts_data %>% filter(!missing) %>% mutate(target = as.integer(target)) %>% select(subject_id, age, trial_id, time, target) ts_data_agg #> # A tibble: 2,560 × 4 #> subject_id age time prop #> #> 1 1413 Older 0 0.7 #> 2 1413 Older 25 0.733 #> 3 1413 Older 50 0.7 #> 4 1413 Older 75 0.7 #> 5 1413 Older 100 0.677 #> 6 1413 Older 125 0.677 #> 7 1413 Older 150 0.645 #> 8 1413 Older 175 0.645 #> 9 1413 Older 200 0.581 #> 10 1413 Older 225 0.548 #> # ℹ 2,550 more rows library(ggplot2) tutorial_fig <- ggplot(ts_data_agg, aes(x = time, y = prop, color = age)) + stat_smooth(method = \"loess\", span = 0.1, se = TRUE, aes(fill = age), alpha = 0.3) + theme_bw() + labs(y = \"Prop. of Target Looks\") + geom_hline(yintercept = .5, color = \"black\") + geom_vline(xintercept = 0, color = \"black\") + coord_cartesian(ylim = c(0, 1)) + theme( legend.position = \"bottom\", legend.title = element_blank() ) tutorial_fig"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"outline","dir":"Articles","previous_headings":"","what":"Outline","title":"Garrison et al. 2020","text":"case study vignette showcases four features CPA jlmerclusterperm: Prepping data CPA using make_jlmer_spec() CPA one go clusterpermute() modular, step--step approach CPA Using logistic regression trial-level data 1s 0s Load package start Julia instance jlmerclusterperm_setup() proceeding.","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"a-prepping-a-specification-object","dir":"Articles","previous_headings":"","what":"A) Prepping a specification object","title":"Garrison et al. 2020","text":"Conducting cluster-based permutation analysis jlmerclusterm starts creating special specification object, compiles information necessary CPA. function make_jlmer_spec() returns specification object class minimally requires two arguments: formula: R regression model formula data: data frame example, model prop using age predictor ts_data_agg data: printed display includes four pieces information specification object: Formula: model formula Julia syntax. , looks similar formula provided, contrasts “spelled ” (age becomes ageOlder) Predictors: list predictors form ( original : expanded ). case, one predictor age expanded ageOlder ’s treatment coded “Younger” reference level. Groupings: specified call make_jlmer_spec(). empty simple_spec provided formula data. Data: transformed dataframe whose columns terms expanded formula. Note default dummy coding applied discrete variable age - now ageOlder 0 “Younger” 1 “Older”. start bare specification object , lacks parts CPA, can use jlmerclusterperm’s simple interface Julia regression models. function jlmer() takes specification object input returns Julia regression model. model represents “global” model fitted entire data, collapsing across time. equivalent lm() model specifications: raise tangent jlmer() recommend sanity-checking design output global model prior using model specifications CPA. regression models fit Julia “hood”, want make sure output comparable expect R. global model can also tell upper bound model complexity. example, global model singular fit, time-wise models fitted subsets data likely . Functions CPA also take object require information time (calculate time-wise statistics) subject/trial (bootstrapped permutation). ts_data_agg data trial-level information summarized subject, just leave . trial argument can omitted object time sample/bin uniquely identified subject. Otherwise, trial column data uniquely identifies time values within subject (example, column items counterbalanced designed participant sees every item one variants).","code":"simple_spec <- make_jlmer_spec(formula = prop ~ 1 + age, data = ts_data_agg) simple_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: prop ~ 1 + ageOlder #> Predictors: #> age: ageOlder #> Data: #> # A tibble: 2,560 × 2 #> prop ageOlder #> #> 1 0.7 1 #> 2 0.733 1 #> 3 0.7 1 #> # ℹ 2,557 more rows #> ──────────────────────────────────────────────────────────────────────────────── global_jmod <- jlmer(simple_spec) tidy(global_jmod) #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.523 0.00387 135. 0 #> 2 ageOlder 0.0984 0.00548 18.0 3.45e-72 library(broom) # for the `tidy()` method for `lm()` global_rmod <- lm(formula = prop ~ 1 + age, data = ts_data_agg) tidy(global_rmod) #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.523 0.00387 135. 0 #> 2 ageOlder 0.0984 0.00548 18.0 4.45e-68 ts_data_agg_spec <- make_jlmer_spec( formula = prop ~ 1 + age, data = ts_data_agg, subject = \"subject_id\", time = \"time\", trial = NULL # This is the default value ) ts_data_agg_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: prop ~ 1 + ageOlder #> Predictors: #> age: ageOlder #> Groupings: #> Subject: subject_id #> Trial: #> Time: time #> Data: #> # A tibble: 2,560 × 4 #> prop ageOlder subject_id time #> #> 1 0.7 1 1413 0 #> 2 0.733 1 1413 25 #> 3 0.7 1 1413 50 #> # ℹ 2,557 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"b-cpa-with-clusterpermute","dir":"Articles","previous_headings":"","what":"B) CPA with clusterpermute()","title":"Garrison et al. 2020","text":"clusterpermute() function runs CPA one go, using information object. addition specification object, must also supply (t-value) threshold number simulations run: result list two elements: object holds information distribution bootstrapped cluster-mass statistics random permutations data: object holds per-predictor information clusters detected, p-values significance testing null: output explained next section, presented purposes simply note results similar reported Katie Von Holzen’s tutorial, copied . minor numerical differences due stochastic nature permutation test ’re comparing t-tests R regression models Julia.","code":"CPA_agg <- clusterpermute( ts_data_agg_spec, threshold = 2, nsim = 100, progress = FALSE # Suppress printing of progress for vignette ) CPA_agg #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> ageOlder (n = 100) #> Mean (SD): -0.844 (29.71) #> Coverage intervals: 95% [-46.664, 71.698] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.0396) #> [2675, 2775]: 11.017 (p=0.4356) #> [3500, 3900]: 46.295 (p=0.0990) #> ──────────────────────────────────────────────────────────────────────────────── #> ## Summary of Clusters ====== #> ## Cluster Direction SumStatistic StartTime EndTime Probability #> ## 1 1 Positive 74.723687 850 1550 0.051 #> ## 2 2 Positive 6.905622 2725 2800 0.405 #> ## 3 3 Positive 46.294644 3500 3925 0.111"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"c-a-step-by-step-approach","dir":"Articles","previous_headings":"","what":"C) A step-by-step approach","title":"Garrison et al. 2020","text":"clusterpermute() function showcased consists five parts, also standalone functions package: compute_timewise_statistics() extract_empirical_clusters() permute_timewise_statistics() extract_null_cluster_dists() calculate_clusters_pvalues() functions correspond algorithmic steps statistical test. clusterpermute() obviates need think individual components, knowing gives greater flexibility procedure. section walks CPA step--step using functions.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"empirical-clusters","dir":"Articles","previous_headings":"C) A step-by-step approach","what":"1) Empirical clusters","title":"Garrison et al. 2020","text":"compute_timewise_statistics() function takes object returns timewise statistics form matrix row predictor column time point: example see 25ms, t-value age -0.127. consistent figure tutorial replicated show two lines mostly overlapping. Just demonstration purposes, use to_jlmer() quickly fit model using formula just data 25ms. see t-value age model exactly saw : construct empirical clusters timewise statistics. Clusters defined continuous span statistics sign whose absolute value passes certain threshold. extract_empirical_clusters() function takes timewise statistics threshold just : t-value threshold 2, detect three clusters data main effect age (ageOlder). numbers brackets show span cluster number right show sum t-values cluster (.k.. cluster-mass statistic). can collect object data frame tidy(): add figure : point know whether size clusters observe (un)likely emerge chance. calculate probability, need simulate null distribution.","code":"empirical_statistics <- compute_timewise_statistics(ts_data_agg_spec) dim(empirical_statistics) #> [1] 1 160 empirical_statistics[1, 1:5, drop = FALSE] # First 5 time points #> Time #> Predictor 0 25 50 75 100 #> ageOlder 0.1166025 -0.1273179 -0.2387778 -0.4022095 -0.2804397 to_jlmer(prop ~ 1 + age, data = filter(ts_data_agg, time == 25)) %>% tidy() %>% filter(term == \"ageOlder\") %>% pull(statistic) #> [1] -0.1273179 empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 #> [2675, 2775]: 11.017 #> [3500, 3900]: 46.295 #> ──────────────────────────────────────────────────────────────────────────────── empirical_clusters_df <- tidy(empirical_clusters) empirical_clusters_df #> # A tibble: 3 × 6 #> predictor id start end length sum_statistic #> #> 1 ageOlder 1 850 1525 28 74.7 #> 2 ageOlder 2 2675 2775 5 11.0 #> 3 ageOlder 3 3500 3900 17 46.3 tutorial_fig + geom_segment( aes(x = start, xend = end, y = -Inf, yend = -Inf), linewidth = 10, color = \"steelblue\", inherit.aes = FALSE, data = empirical_clusters_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"Cluster\", id) ), vjust = -2, inherit.aes = FALSE, data = empirical_clusters_df )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"null-distribution","dir":"Articles","previous_headings":"C) A step-by-step approach","what":"2) Null distribution","title":"Garrison et al. 2020","text":"permute_timewise_statistics() function takes specification object number bootstrapped permutation data simulate (via nsim argument). returned 3-dimensional array like output compute_timewise_statistics() except additional dimension representing simulation: permutation algorithm preserves grouping structures data specified specification object. example, testing -subject predictor like age, age groups participants random swapped preserving temporal structure data. null distribution collection largest cluster-mass statistic simulated data. clusters detected simulation, contributes cluster-mass zero null. use extract_null_cluster_dists() construct null distribution, using threshold value 2 allow comparison empirical clusters: printed, returned object shows summary statistics. can use tidy() collect samples null data frame: null distribution can plotted like : Now left test probability observing cluster-mass statistics extreme detected clusters empirical_clusters, using null_cluster_dists reference.","code":"null_statistics <- permute_timewise_statistics(ts_data_agg_spec, nsim = 100) # simulation by time by predictor dim(null_statistics) #> [1] 100 160 1 # First 5 time points of the first three simulations null_statistics[1:3, 1:5, 1, drop = FALSE] #> , , Predictor = ageOlder #> #> Time #> Sim 0 25 50 75 100 #> 001 -0.1037696 0.1387838 0.4583545 0.4337538 0.69136613 #> 002 1.4358330 1.2745381 1.2985888 0.9111359 1.16401311 #> 003 0.2545470 0.3885999 0.3374939 0.3200497 0.05603617 null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> ageOlder (n = 100) #> Mean (SD): -6.150 (40.29) #> Coverage intervals: 95% [-114.564, 74.425] #> ──────────────────────────────────────────────────────────────────────────────── null_cluster_dists_df <- tidy(null_cluster_dists) null_cluster_dists_df #> # A tibble: 100 × 6 #> predictor start end length sum_statistic sim #> #> 1 ageOlder NA NA NA 0 001 #> 2 ageOlder NA NA NA 0 002 #> 3 ageOlder 2175 2225 3 -6.37 003 #> 4 ageOlder NA NA NA 0 004 #> 5 ageOlder 3425 3475 3 -6.67 005 #> 6 ageOlder NA NA NA 0 006 #> 7 ageOlder NA NA NA 0 007 #> 8 ageOlder NA NA NA 0 008 #> 9 ageOlder NA NA NA 0 009 #> 10 ageOlder 1525 2475 39 -114. 010 #> # ℹ 90 more rows plot( density(null_cluster_dists_df$sum_statistic), main = \"Null distribution of cluster-mass statistics\" )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"significance-test","dir":"Articles","previous_headings":"C) A step-by-step approach","what":"3) Significance test","title":"Garrison et al. 2020","text":"calculate_clusters_pvalues() function computes p-value empirical cluster returns augmented object: , p-values slightly different test stochastic. separate vignette covers issues replicability, seed, rng. collected tidy(), p-values added column: Note reporting pvalues, ’s customary add 1 numerator denominator. observed data, however unlikely, one particular arrangement (permutation) data. controlled via add1 argument - false default calculate_clusters_pvalues() true clusterpermute(). now come full circle: pvalue-augmented object, along object, also clusterpermute() function returned .","code":"empirical_clusters_tested <- calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) empirical_clusters_tested #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.1100) #> [2675, 2775]: 11.017 (p=0.4000) #> [3500, 3900]: 46.295 (p=0.1700) #> ──────────────────────────────────────────────────────────────────────────────── tidy(empirical_clusters_tested) #> # A tibble: 3 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 ageOlder 1 850 1525 28 74.7 0.11 #> 2 ageOlder 2 2675 2775 5 11.0 0.4 #> 3 ageOlder 3 3500 3900 17 46.3 0.17 calculate_clusters_pvalues(empirical_clusters, null_cluster_dists, add1 = TRUE) #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.1188) #> [2675, 2775]: 11.017 (p=0.4059) #> [3500, 3900]: 46.295 (p=0.1782) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"aggregated-vs--trial-level-data","dir":"Articles","previous_headings":"D) As logistic regression","what":"Aggregated vs. trial-level data","title":"Garrison et al. 2020","text":"data using far aggregated data collapsed across trials within subject. words, exactly one measure proportion looks target per time point within subject: possible also fit logistic regression model trial-level measure “hits” (1s 0s) target area interest. un-aggregated, trial-level data comes data preparation code top vignette. demonstrate relationship two data sets: specification object fitting logistic regression models trial-level data requires two changes original: formula must predict binary target variable, instead prop trial grouping must specified uniquely identify time point within subject following appropriate object CPA using logistic regression: , sanity check global model using family = \"binomial\": estimates similar global_jmod previous section, log-odds.","code":"nrow(ts_data_agg) #> [1] 2560 ts_data_agg %>% distinct(subject_id, time) %>% nrow() #> [1] 2560 ts_data_trials #> # A tibble: 62,466 × 5 #> subject_id age trial_id time target #> #> 1 1413 Older 2659 0 1 #> 2 1413 Older 2659 25 1 #> 3 1413 Older 2659 50 1 #> 4 1413 Older 2659 75 1 #> 5 1413 Older 2659 100 1 #> 6 1413 Older 2659 125 1 #> 7 1413 Older 2659 150 1 #> 8 1413 Older 2659 175 1 #> 9 1413 Older 2659 200 1 #> 10 1413 Older 2659 225 1 #> # ℹ 62,456 more rows identical( ts_data_agg, ts_data_trials %>% group_by(subject_id, age, time) %>% summarize(prop = mean(target), .groups = \"drop\") ) #> [1] TRUE ts_data_trial_spec <- make_jlmer_spec( formula = target ~ 1 + age, data = ts_data_trials, subject = \"subject_id\", trial = \"trial_id\", time = \"time\" ) ts_data_trial_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: target ~ 1 + ageOlder #> Predictors: #> age: ageOlder #> Groupings: #> Subject: subject_id #> Trial: trial_id #> Time: time #> Data: #> # A tibble: 62,466 × 5 #> target ageOlder subject_id trial_id time #> #> 1 1 1 1413 2659 0 #> 2 1 1 1413 2659 25 #> 3 1 1 1413 2659 50 #> # ℹ 62,463 more rows #> ──────────────────────────────────────────────────────────────────────────────── global_jmod_binom <- jlmer(ts_data_trial_spec, family = \"binomial\") tidy(global_jmod_binom) #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.113 0.0121 9.31 1.24e- 20 #> 2 ageOlder 0.392 0.0164 23.9 1.08e-126"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"repeating-the-cpa-with-logistic-regression","dir":"Articles","previous_headings":"D) As logistic regression","what":"Repeating the CPA with logistic regression","title":"Garrison et al. 2020","text":"repeat procedure use new specification object set family = \"binomial\": Instead 3 clusters, now detect 5. can annotate detected clusters time series data : differences see aggregated vs. trial-level data completely expected follow straightforwardly following two facts: cluster-mass statistic derived size t-values time point, effect size. virtue unaggregated ts_data_trials data number observations, t-values overall inflated difference means (also see output global_jmod vs. global_jmod_binom well). Whereas aggregated data reflects complete pooling trial-level variation, binomial model trial-level data performs pooling treats trial independent. Thus, certain item experiment show outlier effect one time point, example, strongly influence mass cluster spanning given time point. first issue (inflated statistics) trivial just takes matter adjusting threshold. fact, good “stretching ” t-statistics, get higher resolution shape timewise statistics. However. second issue (pooling) troubling can warp fundamental shape timewise statistics. diagnose factors visualizing timewise statistics. also plot timewise statistics aggregated data comparison: see trial-level binomial model overall inflated t-values, well pronounced “peaks” (specifically maximum 1300ms). middle-ground, partial pooling approaching using mixed-effects models CPA showcased case study vignettes. present purposes, raise threshold value (2.8) gets us close detecting least largest cluster empirical_statistics, spanning 850ms-1525ms: run cluster permutation test threshold just largest cluster: gives us pvalue cluster spanning 850ms 1525ms similar one reported aggregated data: appear converge higher nsim data set. also time 1000-simulation runs showcase speed jlmerclusterperm. Note use set_rng_state(), makes two CPAs generate permutations observed data.","code":"empirical_statistics_binom <- compute_timewise_statistics(ts_data_trial_spec, family = \"binomial\") empirical_clusters_binom <- extract_empirical_clusters(empirical_statistics_binom, threshold = 2) empirical_clusters_binom #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [650, 725]: 8.727 #> [800, 1575]: 112.199 #> [1775, 1875]: 12.764 #> [2600, 2825]: 23.733 #> [3475, 3975]: 62.714 #> ──────────────────────────────────────────────────────────────────────────────── empirical_clusters_binom_df <- tidy(empirical_clusters_binom) tutorial_fig + geom_segment( aes(x = start, xend = end, y = -Inf, yend = -Inf), linewidth = 10, color = \"steelblue\", inherit.aes = FALSE, data = empirical_clusters_binom_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"Cluster\", id) ), vjust = -2, inherit.aes = FALSE, data = empirical_clusters_binom_df ) timewise_statistics_fig <- ggplot(mapping = aes(time, statistic)) + geom_hline(aes(yintercept = 0)) + geom_line( aes(color = \"aggregated\"), linewidth = 1.5, data = tidy(empirical_statistics) ) + geom_line( aes(color = \"trial-level\"), linewidth = 1.5, data = tidy(empirical_statistics_binom) ) timewise_statistics_fig + geom_vline( aes(xintercept = time), data = tidy(empirical_statistics_binom) %>% slice(which.max(abs(statistic))) ) empirical_clusters_binom2 <- extract_empirical_clusters(empirical_statistics_binom, threshold = 2.8) empirical_clusters_binom2 #> ── Empirical clusters (t > 2.8) ──────────────────────── ── #> ageOlder #> [850, 1525]: 102.450 #> [3500, 3875]: 51.365 #> ──────────────────────────────────────────────────────────────────────────────── CPA_trial <- clusterpermute( ts_data_trial_spec, family = \"binomial\", threshold = 2.8, nsim = 100, top_n = 1, # Just test the largest cluster progress = FALSE # Suppress printing of progress for vignette ) CPA_trial$empirical_clusters #> ── Empirical clusters (t > 2.8) ──────────────────────── ── #> ageOlder #> [850, 1525]: 102.450 (p=0.0297) #> ──────────────────────────────────────────────────────────────────────────────── CPA_agg$empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.0396) #> [2675, 2775]: 11.017 (p=0.4356) #> [3500, 3900]: 46.295 (p=0.0990) #> ──────────────────────────────────────────────────────────────────────────────── system.time({ set_rng_state(123L) CPA_agg_1000 <- clusterpermute( ts_data_agg_spec, threshold = 2, nsim = 1000, top_n = 1, progress = FALSE ) }) #> user system elapsed #> 0.00 0.07 5.63 CPA_agg_1000 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> ageOlder (n = 1000) #> Mean (SD): 0.078 (40.29) #> Coverage intervals: 95% [-76.748, 82.717] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.0629) #> ──────────────────────────────────────────────────────────────────────────────── system.time({ set_rng_state(123L) CPA_trial_1000 <- clusterpermute( ts_data_trial_spec, family = \"binomial\", threshold = 2.8, nsim = 1000, top_n = 1, progress = FALSE ) }) #> user system elapsed #> 0.12 0.41 12.87 CPA_trial_1000 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2.8) ──────────── ── #> ageOlder (n = 1000) #> Mean (SD): -0.045 (55.09) #> Coverage intervals: 95% [-119.239, 125.506] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2.8) ──────────────────────── ── #> ageOlder #> [850, 1525]: 102.450 (p=0.0689) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"background","dir":"Articles","previous_headings":"","what":"Background","title":"Geller et al. 2020","text":"data comes lexical decision task using pupilometry (specifically, measure pupil dilation) study differences processing difficulty print vs. cursive script. data available part gazer package Geller, Winn, Mahr, & Mirman 2020. follow Jason Geller’s tutorial used clusterperm package conduct cluster-based permutation analysis (CPA) differences pupil size print cursive conditions. data package (cursive_agg_data) looks slightly different used tutorial, ensure full reproducibility use exact data tutorial used: data comes prepared analysis box. following columns relevant analysis: subject: Unique identifier subjects script: within-participant factor variable (\"cursive\", \"print\") time: continuous measure time 0-2500ms 100ms intervals aggbaseline: response variable representing normalized (“baseline-corrected”) pupil size name cusrive_agg_data variable suggests, data aggregated within subject, collapsing across trials. following reproduces figure tutorial: within-participant predictors like script, permutation algorithm randomly swap labels condition trials within subject. preserves temporal structure trial-level data (swapping trial-level grouping) well subject-level grouping structure (swapping trials across participants).","code":"library(dplyr) cursive_agg_data <- as_tibble(read.csv(\"https://raw.githubusercontent.com/jgeller112/drjasongeller/main/content/blog/2020-07-10-CBPT/blog_data.csv\")) cursive_agg_data #> # A tibble: 2,184 × 5 #> X subject script timebins aggbaseline #> #> 1 1 10b.edf cursive 0 36.9 #> 2 2 10b.edf cursive 100 27.5 #> 3 3 10b.edf cursive 200 19.6 #> 4 4 10b.edf cursive 300 12.6 #> 5 5 10b.edf cursive 400 5.60 #> 6 6 10b.edf cursive 500 0.968 #> 7 7 10b.edf cursive 600 -5.98 #> 8 8 10b.edf cursive 700 -5.96 #> 9 9 10b.edf cursive 800 3.01 #> 10 10 10b.edf cursive 900 12.9 #> # ℹ 2,174 more rows library(ggplot2) #> Warning: package 'ggplot2' was built under R version 4.3.1 script_fig <- ggplot(cursive_agg_data, aes(timebins, aggbaseline)) + stat_summary(aes(linetype = script), geom = \"line\", linewidth = 1) script_fig"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"outline","dir":"Articles","previous_headings":"","what":"Outline","title":"Geller et al. 2020","text":"case study vignette showcases five features CPA jlmerclusterperm: Prepping data CPA using make_jlmer_spec() clusterpermute() default statistic = \"t\" comparison “t” “chisq” Specifying random effects Contrast coding Load package start Julia instance jlmerclusterperm_setup() proceeding.","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"a-prepping-a-specification-object","dir":"Articles","previous_headings":"","what":"A) Prepping a specification object","title":"Geller et al. 2020","text":"start simple specification object model aggbaseline using script predictor script_fig data: sanity check, fit global model data… …check ’s comparable expect lm(): full specification object CPA must also declare grouping structures present data. rule thumb every observation (row) data must uniquely identified combination columns subject, trial, time. CPA time series data time must always specified, subject must also always specified permutation algorithm respect subject-level grouping data. cursive_agg_data data collapses across trials within subject, column trial. However, leave trial argument unspecified observations uniquely identified subject time alone. instead 2 rows per subject-time combination, one script condition: Therefore need “dummy” column trial distinctly mark “cursive” vs. “script” trials. save new data cursive_agg: makes observations uniquely identified columns subject, trial, time: final specification object looks like following:","code":"simple_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script, data = cursive_agg_data ) simple_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: aggbaseline ~ 1 + scriptprint #> Predictors: #> script: scriptprint #> Data: #> # A tibble: 2,184 × 2 #> aggbaseline scriptprint #> #> 1 36.9 0 #> 2 27.5 0 #> 3 19.6 0 #> # ℹ 2,181 more rows #> ──────────────────────────────────────────────────────────────────────────────── jlmer(simple_spec) #> #> ──────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ──────────────────────────────────────────────────────────────────────── #> (Intercept) 46.7221 2.79622 16.71 <1e-61 41.2417 52.2026 #> scriptprint -10.2582 3.95445 -2.59 0.0095 -18.0088 -2.5076 #> ──────────────────────────────────────────────────────────────────────── summary(lm(formula = aggbaseline ~ 1 + script, data = cursive_agg_data))$coefficients #> Estimate Std. Error t value Pr(>|t|) #> (Intercept) 46.72214 2.796221 16.709029 4.512338e-59 #> scriptprint -10.25818 3.954454 -2.594083 9.547816e-03 cursive_agg_data %>% count(subject, timebins) #> # A tibble: 1,092 × 3 #> subject timebins n #> #> 1 10b.edf 0 2 #> 2 10b.edf 100 2 #> 3 10b.edf 200 2 #> 4 10b.edf 300 2 #> 5 10b.edf 400 2 #> 6 10b.edf 500 2 #> 7 10b.edf 600 2 #> 8 10b.edf 700 2 #> 9 10b.edf 800 2 #> 10 10b.edf 900 2 #> # ℹ 1,082 more rows cursive_agg <- cursive_agg_data %>% mutate(trial_type = paste0(script, \"_agg\")) cursive_agg #> # A tibble: 2,184 × 6 #> X subject script timebins aggbaseline trial_type #> #> 1 1 10b.edf cursive 0 36.9 cursive_agg #> 2 2 10b.edf cursive 100 27.5 cursive_agg #> 3 3 10b.edf cursive 200 19.6 cursive_agg #> 4 4 10b.edf cursive 300 12.6 cursive_agg #> 5 5 10b.edf cursive 400 5.60 cursive_agg #> 6 6 10b.edf cursive 500 0.968 cursive_agg #> 7 7 10b.edf cursive 600 -5.98 cursive_agg #> 8 8 10b.edf cursive 700 -5.96 cursive_agg #> 9 9 10b.edf cursive 800 3.01 cursive_agg #> 10 10 10b.edf cursive 900 12.9 cursive_agg #> # ℹ 2,174 more rows cursive_agg %>% count(subject, timebins, trial_type) %>% distinct(n) #> # A tibble: 1 × 1 #> n #> #> 1 1 cursive_agg_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script, data = cursive_agg, subject = \"subject\", trial = \"trial_type\", time = \"timebins\" ) cursive_agg_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: aggbaseline ~ 1 + scriptprint #> Predictors: #> script: scriptprint #> Groupings: #> Subject: subject #> Trial: trial_type #> Time: timebins #> Data: #> # A tibble: 2,184 × 5 #> aggbaseline scriptprint subject trial_type timebins #> #> 1 36.9 0 10b.edf cursive_agg 0 #> 2 27.5 0 10b.edf cursive_agg 100 #> 3 19.6 0 10b.edf cursive_agg 200 #> # ℹ 2,181 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"b-cpa-with-default-statistic-t","dir":"Articles","previous_headings":"","what":"B) CPA with default statistic = \"t\"","title":"Geller et al. 2020","text":"CPA output original tutorial (using 100 simulations) copied comparison: Using threshold 2 default statistic = \"t\", clusterpermute() returns following: results look different original. due following: cluster-mass statistic (cms) lower. ran clusterpermute() default statistic = \"t\". Different clusters identified. simple model account subject-level variation, whereas original error term formula aggbaseline ~ script + Error(subject). sign cluster reversed. defaults aov() used original tutorial different default contrast coding regression model. now address issue turn, building CPA closely replicate results tutorial.","code":"#> #> effect b0 b1 sign cms p #> #> script 0 100 1 10.121215 0.1584158 #> #> script 900 1000 -1 8.756152 0.1782178 #> #> script 1900 2500 1 82.198279 0.0099010 set_rng_state(123L) clusterpermute( cursive_agg_spec, statistic = \"t\", # Default value spelled out threshold = 2, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> scriptprint (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> scriptprint #> [2000, 2300]: -8.748 (p=0.0099) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"c-comparing-t-vs--chisq","dir":"Articles","previous_headings":"","what":"C) Comparing “t” vs. “chisq”","title":"Geller et al. 2020","text":"original tutorial used clusterperm::cluster_nhds() conduct CPA, fits ANOVA models time. , timewise statistic used chi-squared threshold determined p-value chi-squared statistic. Using chi-squared statistics p-value threshold also supported clusterpermute() using statistic = \"chisq\" (instead default \"t\"): returns cluster now numerically larger cluster-mass statistic, expected (chi-squared asymptotic t^2). , compare shape timewise statistics “t” threshold 2 “chisq” threshold p=0.05: find clusters identified 2000ms-2300ms “t” “chisq”, peaks pronounced “chisq”. differences two inconsequential example, may produce different results cases. chi-squared statistic (jlmerclusterperm computes via likelihood ratio test) often preferred testing single parameters makes less assumptions tend robust (glmmFAQ). goodness--fit tests level predictor model formula, “chisq” less interpretable multi-level predictors (k-1 > 1). teasing apart contribution individual levels multi-level predictor, using statistic = \"t\" appropriate.","code":"set_rng_state(123L) clusterpermute( cursive_agg_spec, statistic = \"chisq\", threshold = 0.05, # Threshold is now the p-value of the chi-squared statistic nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [2000, 2300]: -19.085 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── timewise_ts <- compute_timewise_statistics(cursive_agg_spec, statistic = \"t\") timewise_chisqs <- compute_timewise_statistics(cursive_agg_spec, statistic = \"chisq\") library(ggplot2) timewise_fig <- ggplot(mapping = aes(x = time, y = statistic)) + geom_line( aes(color = \"fixed-t\"), linewidth = 1.5, data = tidy(timewise_ts) ) + geom_line( aes(color = \"fixed-chisq\"), linewidth = 1.5, data = tidy(timewise_chisqs) ) + geom_hline( aes(yintercept = c(-2, 2, qchisq(.95, df = 1), -qchisq(.95, df = 1))), color = rep(c(\"#00BFC4\", \"#F8766D\"), each = 2), linetype = 2 ) + scale_color_manual( values = setNames( c(\"#E69F00\", \"#56B4E9\", \"#009E73\", \"#F0E442\"), c(\"fixed-t\", \"fixed-chisq\", \"re-intercept-chisq\", \"re-max-chisq\") ) ) timewise_fig"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"d-specifying-random-effects","dir":"Articles","previous_headings":"","what":"D) Specifying random effects","title":"Geller et al. 2020","text":"biggest missing component point subject random effects, original tutorial captures via error term Error(subject). isn’t strict equivalent regression, specifying random intercepts subject (1 | subject) gets us close: repeat results original tutorial comparison: now plot timewise statistics random intercept models: point may wonder whether results change much used maximal model random effects structure (1 + script | subject). first create another specification object maximal formula: compute timewise statistics: find chisq statistics maximal model virtually identical parsimonious, intercept-model (two lines overlap): Parsimony incredibly important simulation - jlmerclusterperm fast, model complexity still major bottleneck. adding extra random effect terms (random slope correlation) negligible effects statistic, removing likely inconsequential CPA . show following CPA run, uses maximal model. Notice results identical intercept-re_CPA substantially slower: run maximal mixed model CPA purely demonstration purposes. random effects structure actually unidentifiable given aggregated data, point time 2 observations subject (mean condition). Lastly, back figure data, annotate clusters detected chisq p-value threshold 0.05 random intercept CPA:","code":"cursive_agg_spec_re <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 | subject), data = cursive_agg, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) set_rng_state(123L) system.time({ re_CPA <- clusterpermute( cursive_agg_spec_re, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) }) #> user system elapsed #> 0.03 0.06 13.65 re_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): 1.132 (12.58) #> Coverage intervals: 95% [-21.081, 32.649] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: -9.770 (p=0.1881) #> [900, 1000]: 8.521 (p=0.2475) #> [1900, 2500]: -73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── #> #> effect b0 b1 sign cms p #> #> script 0 100 1 10.121215 0.1584158 #> #> script 900 1000 -1 8.756152 0.1782178 #> #> script 1900 2500 1 82.198279 0.0099010 timewise_chisqs_re <- compute_timewise_statistics(cursive_agg_spec_re, statistic = \"chisq\") timewise_fig_re <- timewise_fig + geom_line( aes(color = \"re-intercept-chisq\"), linewidth = 1.5, data = tidy(timewise_chisqs_re) ) timewise_fig_re cursive_agg_spec_re_max <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 + script | subject), data = cursive_agg, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) timewise_chisqs_re_max <- compute_timewise_statistics(cursive_agg_spec_re_max, statistic = \"chisq\") #> ℹ 1 singular fit (3.85%). timewise_fig_re + geom_line( aes(color = \"re-max-chisq\"), linewidth = 1.5, data = tidy(timewise_chisqs_re_max) ) set_rng_state(123L) system.time({ re_max_CPA <- clusterpermute( cursive_agg_spec_re_max, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) }) #> user system elapsed #> 0.06 0.08 12.04 re_max_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): 1.132 (12.58) #> Coverage intervals: 95% [-21.081, 32.649] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: -9.770 (p=0.1881) #> [900, 1000]: 8.521 (p=0.2475) #> [1900, 2500]: -73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── empirical_clusters_df <- tidy(extract_empirical_clusters(timewise_chisqs_re, threshold = 0.05)) script_fig + geom_segment( aes( x = start, xend = end, y = -Inf, yend = -Inf, color = factor(sign(sum_statistic)) ), linewidth = 10, inherit.aes = FALSE, data = empirical_clusters_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"Cluster\", id) ), vjust = -2, inherit.aes = FALSE, data = empirical_clusters_df ) + labs(color = \"Sign of cluster\")"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"e-contrast-coding","dir":"Articles","previous_headings":"","what":"E) Contrast coding","title":"Geller et al. 2020","text":"last piece puzzle flipped sign effect. Whereas detect negative cluster line cursive line print, original tutorial reports positive cluster (output repeated ): Fixing trivial - just takes different choice contrast. example, can flip levels factor make “print” reference level: also sum-code set “print” -1 “cursive” 1: CPA operates domain timewise statistics (effect size), whether difference “print” “cursive” expressed unit 1 (treatment coding) 2 (contrast coding) bearing CPA. conclude, re-run time 1000-simulation CPA using intercept-model new treatment coding script “print” reference level. Finally, annotating just significant cluster figure:","code":"#> #> effect b0 b1 sign cms p #> #> script 0 100 1 10.121215 0.1584158 #> #> script 900 1000 -1 8.756152 0.1782178 #> #> script 1900 2500 1 82.198279 0.0099010 rev_contrast_df <- cursive_agg rev_contrast_df$script <- factor(rev_contrast_df$script, levels = c(\"print\", \"cursive\")) contrasts(rev_contrast_df$script) #> cursive #> print 0 #> cursive 1 reverse_contrast_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 | subject), data = rev_contrast_df, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) set_rng_state(123L) clusterpermute( reverse_contrast_spec, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): -1.132 (12.58) #> Coverage intervals: 95% [-32.649, 21.081] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: 9.770 (p=0.1881) #> [900, 1000]: -8.521 (p=0.2475) #> [1900, 2500]: 73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── sum_contrast_df <- cursive_agg sum_contrast_df$script <- factor(sum_contrast_df$script) contrasts(sum_contrast_df$script) <- contr.sum(2) contrasts(sum_contrast_df$script) #> [,1] #> cursive 1 #> print -1 sum_contrast_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 | subject), data = sum_contrast_df, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) set_rng_state(123L) clusterpermute( sum_contrast_spec, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): -1.132 (12.58) #> Coverage intervals: 95% [-32.649, 21.081] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: 9.770 (p=0.1881) #> [900, 1000]: -8.521 (p=0.2475) #> [1900, 2500]: 73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── set_rng_state(123L) system.time({ final_CPA <- clusterpermute( reverse_contrast_spec, statistic = \"chisq\", threshold = 0.05, nsim = 1000, progress = FALSE ) }) #> user system elapsed #> 0.05 0.08 5.67 final_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 1000, df = 1) #> Mean (SD): -0.107 (16.50) #> Coverage intervals: 95% [-36.966, 41.530] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: 9.770 (p=0.2108) #> [900, 1000]: -8.521 (p=0.2358) #> [1900, 2500]: 73.440 (p=0.0090) #> ──────────────────────────────────────────────────────────────────────────────── signif_clusters_df <- tidy(final_CPA$empirical_clusters) %>% filter(pvalue < 0.05) signif_clusters_df #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 script 3 1900 2500 7 73.4 0.00899 script_fig + geom_segment( aes(x = start, xend = end, y = -Inf, yend = -Inf), color = \"steelblue\", linewidth = 10, inherit.aes = FALSE, data = signif_clusters_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"p =\", signif(pvalue, 3)) ), vjust = -2, inherit.aes = FALSE, data = signif_clusters_df )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"overview-of-cpa","dir":"Articles","previous_headings":"","what":"Overview of CPA","title":"Introduction to jlmerclusterperm","text":"Cluster-based permutation analysis (CPA) simulation-based, non-parametric statistical test difference groups time series. suitable analyzing densely-sampled time data (EEG eye-tracking research) research hypothesis often specified existence effect (e.g., predicted higher-order cognitive mechanisms) agnostic temporal details effect (precise moment emergence). CPA formalizes two intuitions means difference groups: countable unit difference (.e., cluster) contiguous, uninterrupted span sufficiently large differences time point (determined via threshold). degree extremity cluster (.e., cluster-mass statistic) measure sensitive magnitude difference, variability, sample size (e.g., t-statistic regression). CPA procedure identifies empirical clusters time series data tests significance observed cluster-mass statistics bootstrapped permutations data. bootstrapping procedure samples “null” (via random shuffling condition labels), yielding distribution cluster-mass statistics emerging chance. statistical significance cluster probability observing cluster-mass statistic extreme cluster’s simulated null distribution.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"package-design","dir":"Articles","previous_headings":"","what":"Package design","title":"Introduction to jlmerclusterperm","text":"jlmerclusterperm provides wholesale piecemeal approach conducting CPA. main workhorse function package clusterpermute(), composed five smaller functions called internally succession. smaller functions representing algorithmic steps CPA also exported, allow control procedure (e.g., debugging diagnosing CPA run). See function documentation .","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"organization-of-vignettes","dir":"Articles","previous_headings":"","what":"Organization of vignettes","title":"Introduction to jlmerclusterperm","text":"package vignettes roughly divided two groups: topics case studies. researcher data ready analysis, recommended go case studies first pluck relevant bits desired analysis order increasing complexity, case studies : Garrison et al. 2020, introduces package’s functions running CPA, demonstrating wholesale piecemeal approaches. also compares use linear vs. logistic regression calculation timewise statistics. Geller et al. 2020, demonstrates use random effects regression contrasts CPA specification. also compares use t vs. chisq timewise statistics. de Carvalho et al. 2021, showcases using custom regression contrasts test relationship multiple levels factor. also discusses issues surrounding complexity timewise regression model. topics cover general package features: Tidying output: tidy() ways collecting jlmerclusterperm objects tidy data. Julia interface: interacting Julia objects using JuliaConnectoR package. Reproducibility: using Julia RNG make CPA runs reproducible. Asynchronous CPA: running CPA background process using future package.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"compared-to-other-implementations","dir":"Articles","previous_headings":"","what":"Compared to other implementations","title":"Introduction to jlmerclusterperm","text":"R packages cluster-based permutation analysis, clusterperm, eyetrackingR, permuco, permutes. Compared existing implementations, jlmerclusterperm package optimized CPAs based mixed-effects regression models, suitable typical experimental research data multiple, crossed grouping structures (e.g., subjects items). also package interface individual algorithmic steps CPA. Lastly, thanks Julia back-end, bootstrapping fast even mixed models.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"further-readings","dir":"Articles","previous_headings":"","what":"Further readings","title":"Introduction to jlmerclusterperm","text":"original method paper CPA Maris & Oostenveld 2007 recent overview paper Meyer et al. 2021 (CPA EEG research, applies broadly) critical paper Sassenhagen & Draschkow 2019 DOs DON’Ts interpreting reporting CPA results. review paper eyetracking analysis methods Ito & Knoeferle 2022 compares CPA statistical techniques testing difference groups time series data. JOSS paper permuco package (Frossard & Renaud 2021) references therein. Benedikt V. Ehinger’s blog post, includes detailed graphical explainer CPA procedure.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Julia interface","text":"","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"interacting-with-julia-objects","dir":"Articles","previous_headings":"","what":"Interacting with Julia objects","title":"Julia interface","text":"jlmerclusterperm functions collect Julia objects R objects, except jlmer() to_jlmer() return GLM.jl MixedModels.jl fitted model objects. can use functions JuliaConnectoR interact (pointers ) Julia objects: can also call functions Julia packages already installed. example, Effects.jl marginaleffects/emmeans-like features: Note jlmer() to_lmer() just conveniences sanity checking steps cluster-based permutation test, thus used general interface fitting regression models Julia serious analyses.","code":"jmod <- to_jlmer(Reaction ~ Days + (Days | Subject), lme4::sleepstudy) jmod #> #> Variance components: #> Column Variance Std.Dev. Corr. #> Subject (Intercept) 565.51067 23.78047 #> Days 32.68212 5.71683 +0.08 #> Residual 654.94145 25.59182 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 251.405 6.63226 37.91 <1e-99 #> Days 10.4673 1.50224 6.97 <1e-11 #> ────────────────────────────────────────────────── library(JuliaConnectoR) juliaLet(\"x.rePCA\", x = jmod) #> #> (Subject = [0.5406660682947617, 1.0],) juliaCall(\"issingular\", jmod) #> [1] FALSE juliaEval(\"using Effects\") juliaLet(\"effects(x.namedelements, y)\", x = list(Days = 2:5), y = jmod) #> #> 4×5 DataFrame #> Row │ Days Reaction err lower upper #> │ Int64 Float64 Float64 Float64 Float64 #> ─────┼──────────────────────────────────────────── #> 1 │ 2 272.34 7.09427 265.245 279.434 #> 2 │ 3 282.807 7.70544 275.102 290.512 #> 3 │ 4 293.274 8.55557 284.719 301.83 #> 4 │ 5 303.742 9.58128 294.16 313.323"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"julia-global-options","dir":"Articles","previous_headings":"","what":"Julia global options","title":"Julia interface","text":"functions involved cluster-based permutation analysis print progress bars update real time console. function julia_progress() controls whether show progress bar width printed progress bar.","code":"# Set Julia progress options and save old state old_progress_opts <- julia_progress(show = FALSE, width = 30) old_progress_opts #> $show #> [1] TRUE #> #> $width #> [1] 50 julia_progress() #> $show #> [1] FALSE #> #> $width #> [1] 30 # Restored old state julia_progress(old_progress_opts) identical(old_progress_opts, julia_progress()) #> [1] TRUE # The syntax to reset progress options julia_progress(show = TRUE, width = \"auto\") julia_progress()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"the-julia-environment","dir":"Articles","previous_headings":"","what":"The Julia environment","title":"Julia interface","text":"following packages loaded Julia environment via jlmerclusterperm_setup():","code":"cat(readLines(system.file(\"julia/Project.toml\", package = \"jlmerclusterperm\")), sep = \"\\n\") #> [deps] #> DataFrames = \"a93c6f00-e57d-5684-b7b6-d8193f3e46c0\" #> Distributions = \"31c24e10-a181-5473-b8eb-7969acd0382f\" #> GLM = \"38e38edf-8417-5370-95a0-9cbb8c7f171a\" #> JlmerClusterPerm = \"2a59065a-9564-4903-82dd-0e42ce19d0e1\" #> MixedModels = \"ff71e718-51f3-5ec2-a782-8ffcbfa3c316\" #> ProgressMeter = \"92933f4c-e287-5a05-a399-4b506db050ca\" #> Random = \"9a3f8284-a2c9-5f02-9a11-845980a1fd5c\" #> Random123 = \"74087812-796a-5b5d-8853-05524746bad3\" #> StatsBase = \"2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91\" #> StatsModels = \"3eaba693-59b7-5ba5-a881-562e759f1c8d\" #> Suppressor = \"fd094767-a336-5f1f-9728-57cf17d0bbfb\" #> #> [compat] #> DataFrames = \"1.5\" #> GLM = \"1.8.2\" #> MixedModels = \"4.12\" #> ProgressMeter = \"1.7\" #> Random123 = \"1.6\" #> StatsBase = \"0.33\" #> StatsModels = \"0.7\" #> Suppressor = \"0.2.1\" #> julia = \"1.8\""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Reproducibility","text":"Minimal example data: jlmer specification object used showcase:","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) library(dplyr, warn.conflicts = FALSE) library(ggplot2) chickweights <- as_tibble(ChickWeight) %>% mutate(diet = forcats::fct_collapse(Diet, A = c(1, 3), B = c(2, 4))) %>% filter(Time %in% 0:20) ggplot(chickweights, aes(Time, weight, color = diet)) + stat_summary(fun.args = list(mult = 1.96)) + stat_summary(geom = \"line\") + theme_minimal(base_size = 18) jlmer_spec <- make_jlmer_spec( formula = weight ~ diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) jlmer_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + dietB + (1 | Chick) #> Predictors: #> diet: dietB #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight dietB Chick Time #> #> 1 42 0 1 0 #> 2 51 0 1 2 #> 3 59 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"rng-seed-vs--counter","dir":"Articles","previous_headings":"","what":"RNG seed vs. counter","title":"Reproducibility","text":"CPA powered bootstrapping, inherently stochastic procedure. Reproducibility guaranteed clusterpermute() calls ran seed counter values. properties Julia random number generator used jlmerclusterperm refer two slightly different concepts: seed conceptually equivalent “seed” talked R, set.seed(). seed used RNG’s initialization, locking specific state persists session. jlmerclusterperm, default seed 1 primarily controlled via options(\"jlmerclusterperm.seed\") must set calling jlmerclusterperm_setup(). counter initializes 0 increments value every time RNG consumed produce random number. equivalent concept base R, though ’s handy intuitive way interfacing RNG’s state interactive session. counter can manipulated via functions reset_rng_state(), set_rng_state(), get_rng_state(). rest vignette showcase interacting counter state, useful practice.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"behavior-of-rng-counter","dir":"Articles","previous_headings":"","what":"Behavior of RNG counter","title":"Reproducibility","text":"start fresh session, counter initialized zero. specify seed, ’s one: start simple clusterpermute() 100 simulations. reproducibility profile run seed 1 counter 0: run, counter now incremented 3712: call clusterpermute() now runs new counter value, produces different result. , expected given stochastic nature CPA (thus need run tens thousands simulations reporting results): second run also increments counter state, note ’s exactly amount (possible predict): difference two runs specifically null distribution cluster-mass statistics. 100-simulation samples drawn share underlying (null) distribution, though identical one another (rarely ).","code":"get_rng_state() #> [1] 0 get_rng_seed() #> [1] 1 CPA_1_0 <- clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) CPA_1_0 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 100) #> Mean (SD): 0.724 (5.54) #> Coverage intervals: 95% [-10.847, 12.592] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0198) #> ──────────────────────────────────────────────────────────────────────────────── get_rng_state() #> [1] 3712 CPA_1_3712 <- clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) CPA_1_3712 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 100) #> Mean (SD): -0.618 (8.51) #> Coverage intervals: 95% [-18.328, 19.076] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0792) #> ──────────────────────────────────────────────────────────────────────────────── get_rng_state() #> [1] 7427 ggplot(mapping = aes(sum_statistic)) + geom_density( aes(fill = \"1\"), alpha = .5, data = tidy(CPA_1_0$null_cluster_dists) ) + geom_density( aes(fill = \"3712\"), alpha = .5, data = tidy(CPA_1_3712$null_cluster_dists) ) + guides(fill = guide_legend(\"Counter\")) + theme_classic()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"manipulating-rng-counter","dir":"Articles","previous_headings":"","what":"Manipulating RNG counter","title":"Reproducibility","text":"Using set_rng_counter(), possible reproduce prior run. example, reproduce second clusterpermute() call, set counter back 3712 set_rng_state(): results subsequent run counter state identical previous run: reproduce first CPA ran counter 0, can use set_rng_state() simply use reset_rng_state() alias set_rng_state(0): behavior can observed specific step simulating null distribution well. function corresponding step permute_timewise_statistics() run 100 simulations step counter zero: Running cluster detection algorithm result using extract_null_cluster_dists() returns null cluster-mass distribution identical CPA_1_0:","code":"set_rng_state(3712) get_rng_state() #> [1] 3712 identical( CPA_1_3712, clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) ) #> [1] TRUE reset_rng_state() get_rng_state() #> [1] 0 identical( CPA_1_0, clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) ) #> [1] TRUE reset_rng_state() get_rng_state() #> [1] 0 null_statistics <- permute_timewise_statistics(jlmer_spec, nsim = 100) get_rng_state() #> [1] 3712 null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 1) null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 100) #> Mean (SD): 0.724 (5.54) #> Coverage intervals: 95% [-10.847, 12.592] #> ──────────────────────────────────────────────────────────────────────────────── identical(null_cluster_dists, CPA_1_0$null_cluster_dists) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"technical-details-and-notes","dir":"Articles","previous_headings":"","what":"Technical details and notes","title":"Reproducibility","text":"Julia back-end uses thread-safe, counter-based RNG called Threefry (specifically, Random123.Threefry2x) comes Random123.jl Julia library adaptation original algorithm developed Salmon, Moraes, Dror, Shaw (2011). RNG initialized “rounds” value 20, also default. Threefry used speed stability across Julia versions. Additionally, noted earlier, RNG initializes explicit seed 1. behavior different case R, random seed chosen user every new session. consequence absence explicit seeding via options(\"jlmerclusterperm.seed\"), first run every new session (.e., runs counter value zero) identical. Given interface RNG counter, many unique advantages using different seed manipulating RNG counter. interface RNG seed also provided via set_rng_seed() get_rng_seed(). Lastly, please aware changing seed R session (e.g., via set.seed()) effect jlmerclusterperm functions powered Julia back-end.","code":"# Restart Julia session jlmerclusterperm_setup(restart = TRUE, verbose = FALSE) # RNG seed is again 1 get_rng_seed() #> [1] 1 # So results of the first CPA of the session (with counter = 0) is the same get_rng_state() #> [1] 0 identical( CPA_1_0, clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) ) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"miscellaneous","dir":"Articles","previous_headings":"","what":"Miscellaneous","title":"Reproducibility","text":"may wondering whether two CPAs overlap range counter values also share overlapping samples null. actually case practice shuffling CPA implemented: two colored lines represent two 100-simulation bootstrap permutations differ initial RNG counter value 1. point shape two lines look different - one just shifted form another. formal investigations (e.g., simulating power CPA), ’s course best practice generate random seeds. counter convenience balances interactivity reproducibility.","code":"# 100 simulations of timewise statistics ## First CPA using counter at 1 set_rng_state(1) x <- permute_timewise_statistics(jlmer_spec, nsim = 100) ## Second CPA using counter at 2 set_rng_state(2) y <- permute_timewise_statistics(jlmer_spec, nsim = 100) # The null cluster-mass statistics x_null <- extract_null_cluster_dists(x, threshold = 1) x_null_stats <- tidy(x_null)$sum_statistic y_null <- extract_null_cluster_dists(y, threshold = 1) y_null_stats <- tidy(y_null)$sum_statistic # Relationship between two CPAs with similar initial counters matplot(cbind(x_null_stats, y_null_stats), type = \"b\", pch = 1)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Tidying output","text":"Minimal example data: jlmer specification object used showcase:","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) library(dplyr, warn.conflicts = FALSE) library(ggplot2) chickweights <- as_tibble(ChickWeight) %>% mutate(diet = forcats::fct_collapse(Diet, A = c(1, 3), B = c(2, 4))) %>% filter(Time %in% 0:20) ggplot(chickweights, aes(Time, weight, color = diet)) + stat_summary(fun.args = list(mult = 1.96)) + stat_summary(geom = \"line\") + theme_minimal(base_size = 18) jlmer_spec <- make_jlmer_spec( formula = weight ~ diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) jlmer_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + dietB + (1 | Chick) #> Predictors: #> diet: dietB #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight dietB Chick Time #> #> 1 42 0 1 0 #> 2 51 0 1 2 #> 3 59 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"a-tour-of-tidy-methods","dir":"Articles","previous_headings":"","what":"A tour of tidy() methods","title":"Tidying output","text":"non- objects returned jlmerclusterperm functions tidy() method return data frame representation object.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"timewise_statistics-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":"Functions compute_timewise_statistics() permute_timewise_statistics() return object class , array statistics. Note isn’t stylized print() method class. Instead, recommended use tidy() inspect results data frame statistic represented row. case compute_timewise_statistics(), output matrix (2D array) predictor time. dimension becomes column tidy() data: case compute_timewise_statistics(), output 3D array simulation, time, predictor. , dimension becomes column tidy() data:","code":"empirical_statistics <- compute_timewise_statistics(jlmer_spec) class(empirical_statistics) #> [1] \"timewise_statistics\" dim(empirical_statistics) # predictor x time #> [1] 1 11 tidy(empirical_statistics) #> # A tibble: 11 × 3 #> predictor time statistic #> #> 1 dietB 0 -1.09 #> 2 dietB 2 2.29 #> 3 dietB 4 3.14 #> 4 dietB 6 4.01 #> 5 dietB 8 2.90 #> 6 dietB 10 2.44 #> 7 dietB 12 2.18 #> 8 dietB 14 1.26 #> 9 dietB 16 0.673 #> 10 dietB 18 0.532 #> 11 dietB 20 0.809 null_statistics <- permute_timewise_statistics(jlmer_spec, nsim = 1000) class(null_statistics) #> [1] \"timewise_statistics\" dim(null_statistics) # simulation x time x predictor #> [1] 1000 11 1 tidy(null_statistics) #> # A tibble: 11,000 × 4 #> predictor time statistic sim #> #> 1 dietB 0 -0.309 0001 #> 2 dietB 0 2.10 0002 #> 3 dietB 0 -2.83 0003 #> 4 dietB 0 -0.568 0004 #> 5 dietB 0 2.10 0005 #> 6 dietB 0 -1.09 0006 #> 7 dietB 0 -0.830 0007 #> 8 dietB 0 -0.0515 0008 #> 9 dietB 0 -0.0515 0009 #> 10 dietB 0 0.206 0010 #> # ℹ 10,990 more rows"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"empirical_clusters-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_empirical_clusters() print method defined, makes easier eye: tidy() data frame representation also contains information like length, number time points cluster spans. id variable uniquely identifies cluster within predictor.","code":"empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 1) class(empirical_clusters) #> [1] \"empirical_clusters\" empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 #> ──────────────────────────────────────────────────────────────────────────────── tidy(empirical_clusters) #> # A tibble: 1 × 6 #> predictor id start end length sum_statistic #> #> 1 dietB 1 2 14 7 18.2"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"null_cluster_dists-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_null_cluster_dists() also print method defined, displays summary statistics: null distribution constructed largest cluster-mass statistic observed bootstrapped permutations original data. words, simulation contributes one sample null. tidy() representation, simulation predictor represents row. sum_statistic column represents size largest cluster simulation. clusters found, column zero values start, end, length NA. output tidy(null_statistics) tidy(null_cluster_dists) form relational data. example, can use left_join() find time wise statistics permuted data extreme cluster-mass statistic: Note sum_statistic simply sum statistic values span largest cluster:","code":"null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 1) class(null_cluster_dists) #> [1] \"null_cluster_dists\" null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.014 (7.11) #> Coverage intervals: 95% [-15.360, 15.902] #> ──────────────────────────────────────────────────────────────────────────────── tidy(null_cluster_dists) #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 8 20 7 11.1 0001 #> 2 dietB 6 10 3 3.75 0002 #> 3 dietB 18 20 2 2.07 0003 #> 4 dietB 2 10 5 6.00 0004 #> 5 dietB 4 10 4 5.91 0005 #> 6 dietB NA NA NA 0 0006 #> 7 dietB 8 20 7 12.2 0007 #> 8 dietB 2 8 4 5.29 0008 #> 9 dietB NA NA NA 0 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows largest_permuted_cluster <- tidy(null_cluster_dists) %>% slice(which.max(abs(sum_statistic))) %>% left_join(tidy(null_statistics), by = c(\"predictor\", \"sim\")) largest_permuted_cluster #> # A tibble: 11 × 8 #> predictor start end length sum_statistic sim time statistic #> #> 1 dietB 2 20 10 29.4 0319 0 -1.09 #> 2 dietB 2 20 10 29.4 0319 2 1.67 #> 3 dietB 2 20 10 29.4 0319 4 2.56 #> 4 dietB 2 20 10 29.4 0319 6 2.97 #> 5 dietB 2 20 10 29.4 0319 8 3.47 #> 6 dietB 2 20 10 29.4 0319 10 3.54 #> 7 dietB 2 20 10 29.4 0319 12 3.49 #> 8 dietB 2 20 10 29.4 0319 14 3.54 #> 9 dietB 2 20 10 29.4 0319 16 3.05 #> 10 dietB 2 20 10 29.4 0319 18 2.65 #> 11 dietB 2 20 10 29.4 0319 20 2.48 largest_permuted_cluster %>% filter(between(time, start, end)) %>% summarize(sum_statistic = unique(sum_statistic), sum(statistic)) #> # A tibble: 1 × 2 #> sum_statistic `sum(statistic)` #> #> 1 29.4 29.4"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"in-other-contexts","dir":"Articles","previous_headings":"A tour of tidy() methods","what":"In other contexts","title":"Tidying output","text":"output calculate_clusters_pvalues() just object augmented p-values: available, p-values returned column tidy() data: Note p-value cluster 0.027 27 samples null extreme observed: case one-fell-swoop function clusterpermute(), object object returned list: can tidy() elements list lapply() purrr::map():","code":"tested_clusters <- calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) tested_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0270) #> ──────────────────────────────────────────────────────────────────────────────── tidy(tested_clusters) #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.027 tidy(null_cluster_dists) %>% filter(abs(sum_statistic) >= abs(tidy(tested_clusters)$sum_statistic)) #> # A tibble: 27 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 0 20 11 -18.7 0072 #> 2 dietB 0 20 11 -19.4 0114 #> 3 dietB 6 20 8 19.0 0118 #> 4 dietB 0 20 11 -23.6 0137 #> 5 dietB 4 20 9 -21.7 0142 #> 6 dietB 2 20 10 24.7 0268 #> 7 dietB 4 20 9 22.7 0289 #> 8 dietB 2 20 10 29.4 0319 #> 9 dietB 2 20 10 26.4 0325 #> 10 dietB 4 20 9 26.5 0384 #> # ℹ 17 more rows # Timing the entire process this time system.time({ full_test <- clusterpermute(jlmer_spec, threshold = 1, nsim = 1000, progress = FALSE) }) #> user system elapsed #> 0.02 0.05 2.02 full_test #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.066 (7.61) #> Coverage intervals: 95% [-16.959, 17.216] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0420) #> ──────────────────────────────────────────────────────────────────────────────── lapply(full_test, tidy) #> $null_cluster_dists #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 14 20 4 -5.26 0001 #> 2 dietB 14 20 4 9.22 0002 #> 3 dietB 6 20 8 11.8 0003 #> 4 dietB NA NA NA 0 0004 #> 5 dietB NA NA NA 0 0005 #> 6 dietB 16 20 3 -3.69 0006 #> 7 dietB 14 20 4 6.23 0007 #> 8 dietB NA NA NA 0 0008 #> 9 dietB 12 20 5 5.38 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows #> #> $empirical_clusters #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.0420"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"tidying-julia-model-objects","dir":"Articles","previous_headings":"","what":"Tidying Julia model objects","title":"Tidying output","text":"Functions jlmer() to_jlmer() return object class , JuliaConnectoR pointer Julia model object: can use tidy() glance() return information Julia model data frame: mimic behavior methods broom/broom.mixed defined (g)lm/(g)lmer models: Note Julia MixedModels.jl defaults REML = FALSE, whereas TRUE default lme4.","code":"jlmer_mod <- to_jlmer(weight ~ diet + (1 | Chick), chickweights) class(jlmer_mod) #> [1] \"jlmer_mod\" \"JuliaStructProxy\" \"JuliaProxy\" jlmer_mod #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 305.1809 17.4694 #> Residual 3854.0809 62.0812 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 108.16 4.76331 22.71 <1e-99 #> dietB 12.2644 7.45319 1.65 0.0999 #> ────────────────────────────────────────────────── tidy(jlmer_mod) #> # A tibble: 4 × 7 #> effect group term estimate std.error statistic p.value #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 3.83e-114 #> 2 fixed NA dietB 12.3 7.45 1.65 9.99e- 2 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA NA glance(jlmer_mod) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 533 4 62.1 -2972. 5952. 5969. 5944. 529 library(lme4) library(broom.mixed) lmer_mod <- lmer(weight ~ diet + (1 | Chick), chickweights, REML = FALSE) tidy(lmer_mod) #> # A tibble: 4 × 6 #> effect group term estimate std.error statistic #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 #> 2 fixed NA dietB 12.3 7.45 1.65 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA glance(lmer_mod) #> # A tibble: 1 × 7 #> nobs sigma logLik AIC BIC deviance df.residual #> #> 1 533 62.1 -2972. 5952. 5969. 5944. 529"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Tidying output","text":"Minimal example data: jlmer specification object used vignette:","code":"library(jlmerclusterperm) system.time(jlmerclusterperm_setup(verbose = FALSE)) #> user system elapsed #> 0.03 0.03 24.11 julia_progress(show = FALSE) library(dplyr, warn.conflicts = FALSE) library(ggplot2) chickweights <- as_tibble(ChickWeight) %>% mutate(diet = forcats::fct_collapse(Diet, A = c(1, 3), B = c(2, 4))) %>% filter(Time %in% 0:20) ggplot(chickweights, aes(Time, weight, color = diet)) + stat_summary(fun.args = list(mult = 1.96)) + stat_summary(geom = \"line\") + theme_minimal(base_size = 18) jlmer_spec <- make_jlmer_spec( formula = weight ~ diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) jlmer_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + dietB + (1 | Chick) #> Predictors: #> diet: dietB #> Specials: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight dietB Chick Time #> #> 1 42 0 1 0 #> 2 51 0 1 2 #> 3 59 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"a-tour-of-tidy-methods","dir":"Articles","previous_headings":"","what":"A tour of tidy() methods","title":"Tidying output","text":"non- objects returned jlmerclusterperm functions tidy() method return data frame representation object.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"timewise_statistics-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":"Functions compute_timewise_statistics() permute_timewise_statistics() return object class , array statistics. Note isn’t stylized print() method class. Instead, recommended use tidy() inspect results data frame statistic represented row. case compute_timewise_statistics(), output matrix (2D array) predictor time. dimension becomes column tidy() data: case compute_timewise_statistics(), output 3D array simulation, time, predictor. , dimension becomes column tidy() data:","code":"empirical_statistics <- compute_timewise_statistics(jlmer_spec) class(empirical_statistics) #> [1] \"timewise_statistics\" dim(empirical_statistics) # predictor x time #> [1] 1 11 tidy(empirical_statistics) #> # A tibble: 11 × 3 #> predictor time statistic #> #> 1 dietB 0 -1.09 #> 2 dietB 2 2.29 #> 3 dietB 4 3.14 #> 4 dietB 6 4.01 #> 5 dietB 8 2.90 #> 6 dietB 10 2.44 #> 7 dietB 12 2.18 #> 8 dietB 14 1.26 #> 9 dietB 16 0.673 #> 10 dietB 18 0.532 #> 11 dietB 20 0.809 null_statistics <- permute_timewise_statistics(jlmer_spec, nsim = 1000) class(null_statistics) #> [1] \"timewise_statistics\" dim(null_statistics) # simulation x time x predictor #> [1] 1000 11 1 tidy(null_statistics) #> # A tibble: 11,000 × 4 #> predictor time statistic sim #> #> 1 dietB 0 -0.309 0001 #> 2 dietB 0 2.10 0002 #> 3 dietB 0 -2.83 0003 #> 4 dietB 0 -0.568 0004 #> 5 dietB 0 2.10 0005 #> 6 dietB 0 -1.09 0006 #> 7 dietB 0 -0.830 0007 #> 8 dietB 0 -0.0515 0008 #> 9 dietB 0 -0.0515 0009 #> 10 dietB 0 0.206 0010 #> # ℹ 10,990 more rows"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"empirical_clusters-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_empirical_clusters() print method defined, makes easier eye: tidy() data frame representation also contains information like length, number time points cluster spans. id variable uniquely identifies cluster within predictor.","code":"empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 1) class(empirical_clusters) #> [1] \"empirical_clusters\" empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 #> ──────────────────────────────────────────────────────────────────────────────── tidy(empirical_clusters) #> # A tibble: 1 × 6 #> predictor id start end length sum_statistic #> #> 1 dietB 1 2 14 7 18.2"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"null_cluster_dists-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_null_cluster_dists() also print method defined, displays summary statistics: null distribution constructed largest cluster-mass statistic observed bootstrapped permutations original data. words, simulation contributes one sample null. tidy() representation, simulation predictor represents row. sum_statistic column represents size largest cluster simulation. clusters found, column zero values start, end, length NA. output tidy(null_statistics) tidy(null_cluster_dists) form relational data. example, can use left_join() find time wise statistics permuted data extreme cluster-mass statistic: Note sum_statistic simply sum statistic values span largest cluster:","code":"null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 1) class(null_cluster_dists) #> [1] \"null_cluster_dists\" null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.014 (7.11) #> Coverage intervals: 95% [-15.360, 15.902] #> ──────────────────────────────────────────────────────────────────────────────── tidy(null_cluster_dists) #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 8 20 7 11.1 0001 #> 2 dietB 6 10 3 3.75 0002 #> 3 dietB 18 20 2 2.07 0003 #> 4 dietB 2 10 5 6.00 0004 #> 5 dietB 4 10 4 5.91 0005 #> 6 dietB NA NA NA 0 0006 #> 7 dietB 8 20 7 12.2 0007 #> 8 dietB 2 8 4 5.29 0008 #> 9 dietB NA NA NA 0 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows largest_permuted_cluster <- tidy(null_cluster_dists) %>% slice(which.max(abs(sum_statistic))) %>% left_join(tidy(null_statistics), by = c(\"predictor\", \"sim\")) largest_permuted_cluster #> # A tibble: 11 × 8 #> predictor start end length sum_statistic sim time statistic #> #> 1 dietB 2 20 10 29.4 0319 0 -1.09 #> 2 dietB 2 20 10 29.4 0319 2 1.67 #> 3 dietB 2 20 10 29.4 0319 4 2.56 #> 4 dietB 2 20 10 29.4 0319 6 2.97 #> 5 dietB 2 20 10 29.4 0319 8 3.47 #> 6 dietB 2 20 10 29.4 0319 10 3.54 #> 7 dietB 2 20 10 29.4 0319 12 3.49 #> 8 dietB 2 20 10 29.4 0319 14 3.54 #> 9 dietB 2 20 10 29.4 0319 16 3.05 #> 10 dietB 2 20 10 29.4 0319 18 2.65 #> 11 dietB 2 20 10 29.4 0319 20 2.48 largest_permuted_cluster %>% filter(between(time, start, end)) %>% summarize(sum_statistic = unique(sum_statistic), sum(statistic)) #> # A tibble: 1 × 2 #> sum_statistic `sum(statistic)` #> #> 1 29.4 29.4"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"in-other-contexts","dir":"Articles","previous_headings":"A tour of tidy() methods","what":"In other contexts","title":"Tidying output","text":"output calculate_clusters_pvalues() just object augmented p-values: available, p-values returned column tidy() data: Note p-value cluster 0.027 27 samples null extreme observed: case one-fell-swoop function clusterpermute(), object object returned list: can tidy() elements list lapply() purrr::map():","code":"tested_clusters <- calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) tested_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0270) #> ──────────────────────────────────────────────────────────────────────────────── tidy(tested_clusters) #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.027 tidy(null_cluster_dists) %>% filter(abs(sum_statistic) >= abs(tidy(tested_clusters)$sum_statistic)) #> # A tibble: 27 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 0 20 11 -18.7 0072 #> 2 dietB 0 20 11 -19.4 0114 #> 3 dietB 6 20 8 19.0 0118 #> 4 dietB 0 20 11 -23.6 0137 #> 5 dietB 4 20 9 -21.7 0142 #> 6 dietB 2 20 10 24.7 0268 #> 7 dietB 4 20 9 22.7 0289 #> 8 dietB 2 20 10 29.4 0319 #> 9 dietB 2 20 10 26.4 0325 #> 10 dietB 4 20 9 26.5 0384 #> # ℹ 17 more rows # Timing it for fun this time system.time({ full_test <- clusterpermute(jlmer_spec, threshold = 1, nsim = 1000, progress = FALSE) }) #> user system elapsed #> 0.00 0.01 6.53 full_test$null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.066 (7.61) #> Coverage intervals: 95% [-16.959, 17.216] #> ──────────────────────────────────────────────────────────────────────────────── full_test$empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0420) #> ──────────────────────────────────────────────────────────────────────────────── lapply(full_test, tidy) #> $null_cluster_dists #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 14 20 4 -5.26 0001 #> 2 dietB 14 20 4 9.22 0002 #> 3 dietB 6 20 8 11.8 0003 #> 4 dietB NA NA NA 0 0004 #> 5 dietB NA NA NA 0 0005 #> 6 dietB 16 20 3 -3.69 0006 #> 7 dietB 14 20 4 6.23 0007 #> 8 dietB NA NA NA 0 0008 #> 9 dietB 12 20 5 5.38 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows #> #> $empirical_clusters #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.0420"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"tidying-julia-model-objects","dir":"Articles","previous_headings":"","what":"Tidying Julia model objects","title":"Tidying output","text":"Functions jlmer() to_jlmer() return object class , JuliaConnectoR pointer Julia model object: can use tidy() glance() return information Julia model data frame: mimic behavior methods broom/broom.mixed defined (g)lm/(g)lmer models: Note Julia MixedModels.jl defaults REML = FALSE, whereas TRUE default lme4.","code":"jlmer_mod <- to_jlmer(weight ~ diet + (1 | Chick), chickweights) class(jlmer_mod) #> [1] \"jlmer_mod\" \"JuliaStructProxy\" \"JuliaProxy\" jlmer_mod #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 305.1809 17.4694 #> Residual 3854.0809 62.0812 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 108.16 4.76331 22.71 <1e-99 #> dietB 12.2644 7.45319 1.65 0.0999 #> ────────────────────────────────────────────────── tidy(jlmer_mod) #> # A tibble: 4 × 7 #> effect group term estimate std.error statistic p.value #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 3.83e-114 #> 2 fixed NA dietB 12.3 7.45 1.65 9.99e- 2 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA NA glance(jlmer_mod) #> # A tibble: 1 × 7 #> nobs sigma logLik AIC BIC deviance df.residual #> #> 1 533 62.1 -2972. 5952. 5969. 5944. 529 library(lme4) library(broom.mixed) lmer_mod <- lmer(weight ~ diet + (1 | Chick), chickweights, REML = FALSE) tidy(lmer_mod) #> # A tibble: 4 × 6 #> effect group term estimate std.error statistic #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 #> 2 fixed NA dietB 12.3 7.45 1.65 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA glance(lmer_mod) #> # A tibble: 1 × 7 #> nobs sigma logLik AIC BIC deviance df.residual #> #> 1 533 62.1 -2972. 5952. 5969. 5944. 529"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"June Choe. Author, maintainer, copyright holder.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Choe J (2023). jlmerclusterperm: Cluster-Based Permutation Analysis Densely Sampled Time Data. https://github.com/yjunechoe/jlmerclusterperm, https://yjunechoe.github.io/jlmerclusterperm/.","code":"@Manual{, title = {jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data}, author = {June Choe}, year = {2023}, note = {https://github.com/yjunechoe/jlmerclusterperm, https://yjunechoe.github.io/jlmerclusterperm/}, }"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"our-pledge","dir":"","previous_headings":"","what":"Our Pledge","title":"Contributor Covenant Code of Conduct","text":"members, contributors, leaders pledge make participation community harassment-free experience everyone, regardless age, body size, visible invisible disability, ethnicity, sex characteristics, gender identity expression, level experience, education, socio-economic status, nationality, personal appearance, race, caste, color, religion, sexual identity orientation. pledge act interact ways contribute open, welcoming, diverse, inclusive, healthy community.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"our-standards","dir":"","previous_headings":"","what":"Our Standards","title":"Contributor Covenant Code of Conduct","text":"Examples behavior contributes positive environment community include: Demonstrating empathy kindness toward people respectful differing opinions, viewpoints, experiences Giving gracefully accepting constructive feedback Accepting responsibility apologizing affected mistakes, learning experience Focusing best just us individuals, overall community Examples unacceptable behavior include: use sexualized language imagery, sexual attention advances kind Trolling, insulting derogatory comments, personal political attacks Public private harassment Publishing others’ private information, physical email address, without explicit permission conduct reasonably considered inappropriate professional setting","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"enforcement-responsibilities","dir":"","previous_headings":"","what":"Enforcement Responsibilities","title":"Contributor Covenant Code of Conduct","text":"Community leaders responsible clarifying enforcing standards acceptable behavior take appropriate fair corrective action response behavior deem inappropriate, threatening, offensive, harmful. Community leaders right responsibility remove, edit, reject comments, commits, code, wiki edits, issues, contributions aligned Code Conduct, communicate reasons moderation decisions appropriate.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"scope","dir":"","previous_headings":"","what":"Scope","title":"Contributor Covenant Code of Conduct","text":"Code Conduct applies within community spaces, also applies individual officially representing community public spaces. Examples representing community include using official e-mail address, posting via official social media account, acting appointed representative online offline event.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"enforcement","dir":"","previous_headings":"","what":"Enforcement","title":"Contributor Covenant Code of Conduct","text":"Instances abusive, harassing, otherwise unacceptable behavior may reported community leaders responsible enforcement codeofconduct@rstudio.com. complaints reviewed investigated promptly fairly. community leaders obligated respect privacy security reporter incident.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"enforcement-guidelines","dir":"","previous_headings":"","what":"Enforcement Guidelines","title":"Contributor Covenant Code of Conduct","text":"Community leaders follow Community Impact Guidelines determining consequences action deem violation Code Conduct:","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_1-correction","dir":"","previous_headings":"Enforcement Guidelines","what":"1. Correction","title":"Contributor Covenant Code of Conduct","text":"Community Impact: Use inappropriate language behavior deemed unprofessional unwelcome community. Consequence: private, written warning community leaders, providing clarity around nature violation explanation behavior inappropriate. public apology may requested.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_2-warning","dir":"","previous_headings":"Enforcement Guidelines","what":"2. Warning","title":"Contributor Covenant Code of Conduct","text":"Community Impact: violation single incident series actions. Consequence: warning consequences continued behavior. interaction people involved, including unsolicited interaction enforcing Code Conduct, specified period time. includes avoiding interactions community spaces well external channels like social media. Violating terms may lead temporary permanent ban.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_3-temporary-ban","dir":"","previous_headings":"Enforcement Guidelines","what":"3. Temporary Ban","title":"Contributor Covenant Code of Conduct","text":"Community Impact: serious violation community standards, including sustained inappropriate behavior. Consequence: temporary ban sort interaction public communication community specified period time. public private interaction people involved, including unsolicited interaction enforcing Code Conduct, allowed period. Violating terms may lead permanent ban.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_4-permanent-ban","dir":"","previous_headings":"Enforcement Guidelines","what":"4. Permanent Ban","title":"Contributor Covenant Code of Conduct","text":"Community Impact: Demonstrating pattern violation community standards, including sustained inappropriate behavior, harassment individual, aggression toward disparagement classes individuals. Consequence: permanent ban sort public interaction within community.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"attribution","dir":"","previous_headings":"","what":"Attribution","title":"Contributor Covenant Code of Conduct","text":"Code Conduct adapted Contributor Covenant, version 2.1, available https://www.contributor-covenant.org/version/2/1/code_of_conduct.html. Community Impact Guidelines inspired [Mozilla’s code conduct enforcement ladder][https://github.com/mozilla/inclusion]. answers common questions code conduct, see FAQ https://www.contributor-covenant.org/faq. Translations available https://www.contributor-covenant.org/translations.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":null,"dir":"","previous_headings":"","what":"Contributing to jlmerclusterperm","title":"Contributing to jlmerclusterperm","text":"outlines propose change jlmerclusterperm.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"fixing-typos","dir":"","previous_headings":"","what":"Fixing typos","title":"Contributing to jlmerclusterperm","text":"can fix typos, spelling mistakes, grammatical errors documentation directly using GitHub web interface, long changes made source file. generally means ’ll need edit roxygen2 comments .R, .Rd file. can find .R file generates .Rd reading comment first line.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"bigger-changes","dir":"","previous_headings":"","what":"Bigger changes","title":"Contributing to jlmerclusterperm","text":"want make bigger change, ’s good idea first file issue make sure someone team agrees ’s needed. ’ve found bug, please file issue illustrates bug minimal reprex (also help write unit test, needed).","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"pull-request-process","dir":"","previous_headings":"Bigger changes","what":"Pull request process","title":"Contributing to jlmerclusterperm","text":"Fork package clone onto computer. haven’t done , recommend using usethis::create_from_github(\"yjunechoe/jlmerclusterperm\", fork = TRUE). Install development dependencies devtools::install_dev_deps(), make sure package passes R CMD check running devtools::check(). R CMD check doesn’t pass cleanly, ’s good idea ask help continuing. Create Git branch pull request (PR). recommend using usethis::pr_init(\"brief-description--change\"). Make changes, commit git, create PR running usethis::pr_push(), following prompts browser. title PR briefly describe change. body PR contain Fixes #issue-number. user-facing changes, add bullet top NEWS.md (.e. just first header). Follow style described https://style.tidyverse.org/news.html.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"code-style","dir":"","previous_headings":"Bigger changes","what":"Code style","title":"Contributing to jlmerclusterperm","text":"New code follow tidyverse style guide. can use styler package apply styles, please don’t restyle code nothing PR. use roxygen2, Markdown syntax, documentation. use testthat unit tests. Contributions test cases included easier accept.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"code-of-conduct","dir":"","previous_headings":"","what":"Code of Conduct","title":"Contributing to jlmerclusterperm","text":"Please note jlmerclusterperm project released Contributor Code Conduct. contributing project agree abide terms.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"jlmerclusterperm-","dir":"","previous_headings":"","what":"jlmerclusterperm","title":"jlmerclusterperm","text":"Julia GLM.jl MixedModels.jl based implementation cluster-based permutation test time series data, powered JuliaConnectoR.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"installation-and-usage","dir":"","previous_headings":"","what":"Installation and usage","title":"jlmerclusterperm","text":"Install released version jlmerclusterperm CRAN: install development version GitHub : Using jlmerclusterperm requires prior installation Julia programming language, can downloaded either official website using command line utility juliaup. Julia version >=1.8 required 1.9 preferred substantial speed improvements. using functions jlmerclusterperm, initial setup required via calling jlmerclusterperm_setup(). first call system install necessary dependencies (happens takes around 10-15 minutes). Subsequent calls jlmerclusterperm_setup() incur small overhead around 30 seconds, plus slight delays first-time function calls. pay front start-warm-costs get blazingly-fast functions package. See Get Started page package website background tutorials.","code":"install.packages(\"jlmerclusterperm\") # install.packages(\"remotes\") remotes::install_github(\"yjunechoe/jlmerclusterperm\") # Both lines must be run at the start of each new session library(jlmerclusterperm) jlmerclusterperm_setup()"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"wholesale-cpa-with-clusterpermute","dir":"","previous_headings":"Quick tour of package functionalities","what":"Wholesale CPA with clusterpermute()","title":"jlmerclusterperm","text":"time series data: Preparing specification object make_jlmer_spec(): Cluster-based permutation test clusterpermute(): Including random effects:","code":"chickweights <- ChickWeight chickweights$Time <- as.integer(factor(chickweights$Time)) matplot( tapply(chickweights$weight, chickweights[c(\"Time\", \"Diet\")], mean), type = \"b\", lwd = 3, ylab = \"Weight\", xlab = \"Time\" ) chickweights_spec <- make_jlmer_spec( formula = weight ~ 1 + Diet, data = chickweights, subject = \"Chick\", time = \"Time\" ) chickweights_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> weight Diet2 Diet3 Diet4 Chick Time #> 1 42 0 0 0 1 1 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 3 #> [ reached 'max' / getOption(\"max.print\") -- omitted 575 rows ] #> ──────────────────────────────────────────────────────────────────────────────── set_rng_state(123L) clusterpermute( chickweights_spec, threshold = 2.5, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2.5) ──────────── ── #> Diet2 (n = 100) #> Mean (SD): -0.039 (1.89) #> Coverage intervals: 95% [-2.862, 0.000] #> Diet3 (n = 100) #> Mean (SD): -0.129 (2.02) #> Coverage intervals: 95% [0.000, 0.000] #> Diet4 (n = 100) #> Mean (SD): 0.296 (3.21) #> Coverage intervals: 95% [0.000, 5.797] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.121 (p=0.0495) #> Diet3 #> [3, 12]: 35.769 (p=0.0099) #> Diet4 #> [2, 8]: 32.442 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── chickweights_re_spec <- make_jlmer_spec( formula = weight ~ 1 + Diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) set_rng_state(123L) clusterpermute( chickweights_re_spec, threshold = 2.5, nsim = 100, progress = FALSE )$empirical_clusters #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.387 (p=0.0594) #> Diet3 #> [2, 12]: 39.919 (p=0.0099) #> Diet4 #> [2, 8]: 33.853 (p=0.0099) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"piecemeal-approach-to-cpa","dir":"","previous_headings":"Quick tour of package functionalities","what":"Piecemeal approach to CPA","title":"jlmerclusterperm","text":"Computing time-wise statistics observed data: Identifying empirical clusters: Simulating null distribution: Significance testing cluster-mass statistic: Iterating range threshold values:","code":"empirical_statistics <- compute_timewise_statistics(chickweights_spec) matplot(t(empirical_statistics), type = \"b\", pch = 1, lwd = 3, ylab = \"t-statistic\") abline(h = 2.5, lty = 3) empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2.5) empirical_clusters #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.121 #> Diet3 #> [3, 12]: 35.769 #> Diet4 #> [2, 8]: 32.442 #> ──────────────────────────────────────────────────────────────────────────────── set_rng_state(123L) null_statistics <- permute_timewise_statistics(chickweights_spec, nsim = 100) null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2.5) null_cluster_dists #> ── Null cluster-mass distribution (t > 2.5) ──────────── ── #> Diet2 (n = 100) #> Mean (SD): -0.039 (1.89) #> Coverage intervals: 95% [-2.862, 0.000] #> Diet3 (n = 100) #> Mean (SD): -0.129 (2.02) #> Coverage intervals: 95% [0.000, 0.000] #> Diet4 (n = 100) #> Mean (SD): 0.296 (3.21) #> Coverage intervals: 95% [0.000, 5.797] #> ──────────────────────────────────────────────────────────────────────────────── calculate_clusters_pvalues(empirical_clusters, null_cluster_dists, add1 = TRUE) #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.121 (p=0.0495) #> Diet3 #> [3, 12]: 35.769 (p=0.0099) #> Diet4 #> [2, 8]: 32.442 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── walk_threshold_steps(empirical_statistics, null_statistics, steps = c(2, 2.5, 3)) #> threshold predictor id start end length sum_statistic pvalue #> 1 2.0 Diet2 1 3 5 3 8.496376 0.07920792 #> 2 2.0 Diet3 1 2 12 11 38.216035 0.00990099 #> 3 2.0 Diet4 1 2 12 11 41.651468 0.00990099 #> 4 2.5 Diet2 1 3 4 2 6.121141 0.04950495 #> 5 2.5 Diet3 1 3 12 10 35.768957 0.00990099 #> 6 2.5 Diet4 1 2 8 7 32.442352 0.00990099 #> 31 3.0 Diet3 1 3 5 3 12.719231 0.00990099 #> 21 3.0 Diet3 2 9 12 4 14.037622 0.00990099 #> 41 3.0 Diet4 1 2 7 6 29.659402 0.00990099"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"acknowledgments","dir":"","previous_headings":"","what":"Acknowledgments","title":"jlmerclusterperm","text":"paper Maris & Oostenveld (2007) originally proposed cluster-based permutation analysis. JuliaConnectoR package powering R interface Julia. Julia packages GLM.jl MixedModels.jl fast implementations (mixed effects) regression models. Existing implementations CPA R (permuco, permutes, etc.) whose designs inspired CPA interface jlmerclusterperm.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"citations","dir":"","previous_headings":"","what":"Citations","title":"jlmerclusterperm","text":"use jlmerclusterperm cluster-based permutation test mixed-effects models research, please cite one () following see fit. cite jlmerclusterperm: Choe, J. (2023). jlmerclusterperm: Cluster-Based Permutation Analysis Densely Sampled Time Data. R package version 1.0.4. https://cran.r-project.org/package=jlmerclusterperm. cite cluster-based permutation test: Maris, E., & Oostenveld, R. (2007). Nonparametric statistical testing EEG- MEG-data. Journal Neuroscience Methods, 164, 177–190. doi: 10.1016/j.jneumeth.2007.03.024. cite Julia programming language: Bezanson, J., Edelman, ., Karpinski, S., & Shah, V. B. (2017). Julia: Fresh Approach Numerical Computing. SIAM Review, 59(1), 65–98. doi: 10.1137/141000671. cite GLM.jl MixedModels.jl Julia libraries, consult Zenodo pages: GLM: https://doi.org/10.5281/zenodo.3376013 MixedModels: https://zenodo.org/badge/latestdoi/9106942","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/LICENSE.html","id":null,"dir":"","previous_headings":"","what":"MIT License","title":"MIT License","text":"Copyright (c) 2023 June Choe Permission hereby granted, free charge, person obtaining copy software associated documentation files (“Software”), deal Software without restriction, including without limitation rights use, copy, modify, merge, publish, distribute, sublicense, /sell copies Software, permit persons Software furnished , subject following conditions: copyright notice permission notice shall included copies substantial portions Software. SOFTWARE PROVIDED “”, WITHOUT WARRANTY KIND, EXPRESS IMPLIED, INCLUDING LIMITED WARRANTIES MERCHANTABILITY, FITNESS PARTICULAR PURPOSE NONINFRINGEMENT. EVENT SHALL AUTHORS COPYRIGHT HOLDERS LIABLE CLAIM, DAMAGES LIABILITY, WHETHER ACTION CONTRACT, TORT OTHERWISE, ARISING , CONNECTION SOFTWARE USE DEALINGS SOFTWARE.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"Calculate bootstrapped p-values cluster-mass statistics","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"","code":"calculate_clusters_pvalues( empirical_clusters, null_cluster_dists, add1 = FALSE ) clusters_are_comparable(empirical_clusters, null_cluster_dists, error = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"empirical_clusters empirical_clusters object null_cluster_dists null_cluster_dists object add1 Whether add 1 numerator denominator calculating p-value. Use TRUE effectively count observed statistic part permuted null distribution (recommended larger nsim prior publishing results). error Whether throw error incompatible","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"empirical_clusters object augmented p-values.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Make empirical clusters empirical_statistics <- compute_timewise_statistics(spec) empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 #> Diet3 #> [2, 20]: 34.628 #> Diet4 #> [2, 20]: 39.371 #> ──────────────────────────────────────────────────────────────────────────────── # Make null cluster-mass distribution reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) # Significance test the empirical cluster(s) from each predictor against the simulated null calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0200) #> Diet3 #> [2, 20]: 34.628 (p=0.0000) #> Diet4 #> [2, 20]: 39.371 (p=0.0000) #> ──────────────────────────────────────────────────────────────────────────────── # Set `add1 = TRUE` to normalize by adding 1 to numerator and denominator calculate_clusters_pvalues(empirical_clusters, null_cluster_dists, add1 = TRUE) #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # This sequence of procedures is equivalent to `clusterpermute()` reset_rng_state() clusterpermute(spec, threshold = 2, nsim = 100, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # The empirical clusters and the null cluster-mass distribution must be comparable empirical_clusters2 <- extract_empirical_clusters(empirical_statistics, threshold = 3) # For example, below code errors because thresholds are different (2 vs. 3) try( calculate_clusters_pvalues(empirical_clusters2, null_cluster_dists) ) #> Error in clusters_are_comparable(empirical_clusters, null_cluster_dists, : #> Cluster-mass statistics between empirical and null are not comparable. #> ℹ `empirical_clusters` and `null_cluster_dists` must share the same `statistic` #> and `threshold`. #> ✖ threshold: empirical uses 3 but null uses 2. # Check for compatibility with `clusters_are_comparable()` clusters_are_comparable(empirical_clusters, null_cluster_dists) #> [1] TRUE clusters_are_comparable(empirical_clusters2, null_cluster_dists) #> [1] FALSE # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":null,"dir":"Reference","previous_headings":"","what":"Conduct a cluster-based permutation test — clusterpermute","title":"Conduct a cluster-based permutation test — clusterpermute","text":"Conduct cluster-based permutation test","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Conduct a cluster-based permutation test — clusterpermute","text":"","code":"clusterpermute( jlmer_spec, family = c(\"gaussian\", \"binomial\"), statistic = c(\"t\", \"chisq\"), threshold, nsim = 100L, predictors = NULL, binned = FALSE, top_n = Inf, add1 = TRUE, ..., progress = TRUE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Conduct a cluster-based permutation test — clusterpermute","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". statistic Test statistic calculating cluster mass. Can one \"t\" (default) regression model output \"chisq\" likelihood ratio test (takes twice long calculate). threshold threshold value statistic must pass contribute cluster mass. Interpretation differs choice statistic (): statistic = \"t\", threshold t-value (beta/std.err) regression model. statistic = \"chisq\", threshold p-value chi-squared statistics likelihood ratio tests. nsim Number simulations description predictors (Optional) subset predictors test. Defaults NULL tests predictors. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist. top_n many clusters return, order size cluster-mass statistic. Defaults Inf return detected clusters. add1 Whether add 1 numerator denominator calculating p-value. Use TRUE effectively count observed statistic part permuted null distribution (recommended larger nsim prior publishing results). ... Optional arguments passed Julia model fitting. Defaults fast = TRUE (family = \"binomial\") progress = FALSE. progress Defaults TRUE, prints progress step cluster permutation test.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Conduct a cluster-based permutation test — clusterpermute","text":"list null_cluster_dists empirical_clusters p-values","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Conduct a cluster-based permutation test — clusterpermute","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Should minimally provide `threshold` and `nsim`, in addition to the spec object reset_rng_state() CPA <- clusterpermute(spec, threshold = 2, nsim = 100, progress = FALSE) CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # CPA is a list of `` and `` objects sapply(CPA, class) #> null_cluster_dists empirical_clusters #> \"null_cluster_dists\" \"empirical_clusters\" # You can extract the individual components for further inspection CPA$null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── CPA$empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":null,"dir":"Reference","previous_headings":"","what":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"Tidiers cluster permutation test objects","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"","code":"# S3 method for timewise_statistics tidy(x, ...) # S3 method for empirical_clusters tidy(x, ...) # S3 method for null_cluster_dists tidy(x, ...)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"x object class , , ... Unused","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"data frame","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Method for `` empirical_statistics <- compute_timewise_statistics(spec) class(empirical_statistics) #> [1] \"timewise_statistics\" tidy(empirical_statistics) #> # A tibble: 33 × 3 #> predictor time statistic #> #> 1 Diet2 0 -1.60 #> 2 Diet3 0 -1.37 #> 3 Diet4 0 -0.916 #> 4 Diet2 2 1.67 #> 5 Diet3 2 2.45 #> 6 Diet4 2 3.53 #> 7 Diet2 4 2.60 #> 8 Diet3 4 4.48 #> 9 Diet4 4 6.27 #> 10 Diet2 6 3.52 #> # ℹ 23 more rows reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) class(null_statistics) #> [1] \"timewise_statistics\" tidy(null_statistics) #> # A tibble: 3,300 × 4 #> predictor time statistic sim #> #> 1 Diet2 0 -0.898 001 #> 2 Diet2 0 1.40 002 #> 3 Diet2 0 -1.44 003 #> 4 Diet2 0 -0.449 004 #> 5 Diet2 0 1.38 005 #> 6 Diet2 0 -0.692 006 #> 7 Diet2 0 0.113 007 #> 8 Diet2 0 0.334 008 #> 9 Diet2 0 0.558 009 #> 10 Diet2 0 -0.449 010 #> # ℹ 3,290 more rows # Method for `` empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) class(empirical_clusters) #> [1] \"empirical_clusters\" tidy(empirical_clusters) #> # A tibble: 3 × 6 #> predictor id start end length sum_statistic #> #> 1 Diet2 1 4 8 3 8.50 #> 2 Diet3 1 2 20 10 34.6 #> 3 Diet4 1 2 20 10 39.4 # Method for `` null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) class(null_cluster_dists) #> [1] \"null_cluster_dists\" tidy(null_cluster_dists) #> # A tibble: 300 × 6 #> predictor start end length sum_statistic sim #> #> 1 Diet2 10 20 6 14.8 001 #> 2 Diet2 NA NA NA 0 002 #> 3 Diet2 NA NA NA 0 003 #> 4 Diet2 NA NA NA 0 004 #> 5 Diet2 NA NA NA 0 005 #> 6 Diet2 NA NA NA 0 006 #> 7 Diet2 NA NA NA 0 007 #> 8 Diet2 NA NA NA 0 008 #> 9 Diet2 NA NA NA 0 009 #> 10 Diet2 NA NA NA 0 010 #> # ℹ 290 more rows # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"Fit Julia regression models time point time series data","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"","code":"compute_timewise_statistics( jlmer_spec, family = c(\"gaussian\", \"binomial\"), statistic = c(\"t\", \"chisq\"), ... )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". statistic Test statistic calculating cluster mass. Can one \"t\" (default) regression model output \"chisq\" likelihood ratio test (takes twice long calculate). ... Optional arguments passed Julia model fitting. Defaults fast = TRUE (family = \"binomial\") progress = FALSE.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"predictor--time matrix cluster statistics, class timewise_statistics.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Predictor x Time matrix of t-statistics from regression output empirical_statistics <- compute_timewise_statistics(spec) round(empirical_statistics, 2) #> Time #> Predictor 0 2 4 6 8 10 12 14 16 18 20 #> Diet2 -1.60 1.67 2.60 3.52 2.38 1.91 1.94 1.34 1.16 1.40 1.51 #> Diet3 -1.37 2.45 4.48 4.54 3.70 2.97 3.06 2.98 3.05 3.60 3.80 #> Diet4 -0.92 3.53 6.27 7.00 5.12 4.07 3.66 2.78 2.16 2.13 2.64 # Collect as dataframe with `tidy()` empirical_statistics_df <- tidy(empirical_statistics) empirical_statistics_df #> # A tibble: 33 × 3 #> predictor time statistic #> #> 1 Diet2 0 -1.60 #> 2 Diet3 0 -1.37 #> 3 Diet4 0 -0.916 #> 4 Diet2 2 1.67 #> 5 Diet3 2 2.45 #> 6 Diet4 2 3.53 #> 7 Diet2 4 2.60 #> 8 Diet3 4 4.48 #> 9 Diet4 4 6.27 #> 10 Diet2 6 3.52 #> # ℹ 23 more rows # Timewise statistics are from regression models fitted to each time point # - Note the identical statistics at `Time == 0` empirical_statistics_df %>% filter(time == 0) #> # A tibble: 3 × 3 #> predictor time statistic #> #> 1 Diet2 0 -1.60 #> 2 Diet3 0 -1.37 #> 3 Diet4 0 -0.916 to_jlmer(weight ~ 1 + Diet, filter(ChickWeight, Time == 0)) %>% tidy() %>% select(term, statistic) #> # A tibble: 4 × 2 #> term statistic #> #> 1 (Intercept) 164. #> 2 Diet2 -1.60 #> 3 Diet3 -1.37 #> 4 Diet4 -0.916 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":null,"dir":"Reference","previous_headings":"","what":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"Detect largest clusters time sequence predictor statistics","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"","code":"extract_empirical_clusters( empirical_statistics, threshold, binned = FALSE, top_n = Inf )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"empirical_statistics predictor--time matrix empirical timewise statistics. threshold threshold value statistic must pass contribute cluster mass. Interpretation differs choice statistic (): statistic = \"t\", threshold t-value (beta/std.err) regression model. statistic = \"chisq\", threshold p-value chi-squared statistics likelihood ratio tests. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist. top_n many clusters return, order size cluster-mass statistic. Defaults Inf return detected clusters.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"empirical_clusters object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Empirical clusters are derived from the timewise statistics empirical_statistics <- compute_timewise_statistics(spec) empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 #> Diet3 #> [2, 20]: 34.628 #> Diet4 #> [2, 20]: 39.371 #> ──────────────────────────────────────────────────────────────────────────────── # Collect as dataframe with `tidy()` empirical_clusters_df <- tidy(empirical_clusters) empirical_clusters_df #> # A tibble: 3 × 6 #> predictor id start end length sum_statistic #> #> 1 Diet2 1 4 8 3 8.50 #> 2 Diet3 1 2 20 10 34.6 #> 3 Diet4 1 2 20 10 39.4 # Changing the `threshold` value identifies different clusters extract_empirical_clusters(empirical_statistics, threshold = 1) #> ── Empirical clusters (t > 1) ────────────────────────── ── #> Diet2 #> [2, 20]: 19.427 #> Diet3 #> [2, 20]: 34.628 #> Diet4 #> [2, 20]: 39.371 #> ──────────────────────────────────────────────────────────────────────────────── # A predictor can have zero or multiple clusters associated with it extract_empirical_clusters(empirical_statistics, threshold = 3) #> ── Empirical clusters (t > 3) ────────────────────────── ── #> Diet3 #> [4, 8]: 12.719 #> [16, 20]: 10.449 #> Diet4 #> [2, 12]: 29.659 #> ──────────────────────────────────────────────────────────────────────────────── #> ! No clusters found for Diet2 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":null,"dir":"Reference","previous_headings":"","what":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"Construct null distribution cluster-mass statistics","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"","code":"extract_null_cluster_dists(null_statistics, threshold, binned = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"null_statistics simulation--time--predictor 3D array null (permuted) timewise statistics. threshold threshold value statistic must pass contribute cluster mass. Interpretation differs choice statistic (): statistic = \"t\", threshold t-value (beta/std.err) regression model. statistic = \"chisq\", threshold p-value chi-squared statistics likelihood ratio tests. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"null_cluster_dists object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Null cluster-mass distributions are derived from the permuted timewise statistics reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── # Collect as dataframe with `tidy()` # - Each simulation contributes one (largest) cluster-mass statistic to the null # - When no clusters are found, the `sum_statistic` value is zero null_cluster_dists_df <- tidy(null_cluster_dists) null_cluster_dists_df #> # A tibble: 300 × 6 #> predictor start end length sum_statistic sim #> #> 1 Diet2 10 20 6 14.8 001 #> 2 Diet2 NA NA NA 0 002 #> 3 Diet2 NA NA NA 0 003 #> 4 Diet2 NA NA NA 0 004 #> 5 Diet2 NA NA NA 0 005 #> 6 Diet2 NA NA NA 0 006 #> 7 Diet2 NA NA NA 0 007 #> 8 Diet2 NA NA NA 0 008 #> 9 Diet2 NA NA NA 0 009 #> 10 Diet2 NA NA NA 0 010 #> # ℹ 290 more rows # Changing the `threshold` value changes the shape of the null extract_null_cluster_dists(null_statistics, threshold = 1) #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.565 (5.52) #> Coverage intervals: 95% [-9.681, 12.061] #> Diet3 (n = 100) #> Mean (SD): 0.089 (6.27) #> Coverage intervals: 95% [-15.847, 12.185] #> Diet4 (n = 100) #> Mean (SD): 0.715 (5.23) #> Coverage intervals: 95% [-7.843, 11.728] #> ──────────────────────────────────────────────────────────────────────────────── extract_null_cluster_dists(null_statistics, threshold = 3) #> ── Null cluster-mass distribution (t > 3) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> Diet3 (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> Diet4 (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit a Julia regression model using jlmer specifications — jlmer","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"Fit Julia regression model using jlmer specifications","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"","code":"jlmer(jlmer_spec, family = c(\"gaussian\", \"binomial\"), ..., progress = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". ... Optional arguments passed Julia model fitting. progress TRUE, prints timing iterations.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"jlmer_mod object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } jlmerclusterperm_setup(verbose = FALSE) # Fitting a regression model with a specification object spec <- make_jlmer_spec(weight ~ 1 + Diet, ChickWeight) jlmer(spec) #> #> ──────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ──────────────────────────────────────────────────────────────────────── #> (Intercept) 102.645 4.67395 21.96 <1e-99 93.4847 111.806 #> Diet2 19.9712 7.86744 2.54 0.0111 4.55132 35.3911 #> Diet3 40.3045 7.86744 5.12 <1e-06 24.8847 55.7244 #> Diet4 32.6173 7.91046 4.12 <1e-04 17.113 48.1215 #> ──────────────────────────────────────────────────────────────────────── # `lm()` equivalent summary(lm(weight ~ 1 + Diet, ChickWeight))$coef #> Estimate Std. Error t value Pr(>|t|) #> (Intercept) 102.64545 4.673954 21.961161 4.712762e-78 #> Diet2 19.97121 7.867437 2.538465 1.139708e-02 #> Diet3 40.30455 7.867437 5.122958 4.113938e-07 #> Diet4 32.61726 7.910461 4.123307 4.286352e-05 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm-package.html","id":null,"dir":"Reference","previous_headings":"","what":"jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data — jlmerclusterperm-package","title":"jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data — jlmerclusterperm-package","text":"implementation fast cluster-based permutation analysis (CPA) densely-sampled time data developed Maris & Oostenveld, 2007 doi:10.1016/j.jneumeth.2007.03.024 . Supports (generalized, mixed-effects) regression models calculation timewise statistics. Provides wholesale piecemeal interface CPA procedure emphasis interpretability diagnostics. Integrates 'Julia' libraries 'MixedModels.jl' 'GLM.jl' performance improvements, additional functionalities interfacing 'Julia' 'R' powered 'JuliaConnectoR' package.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data — jlmerclusterperm-package","text":"Maintainer: June Choe jchoe001@gmail.com (ORCID) [copyright holder]","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":null,"dir":"Reference","previous_headings":"","what":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"Initial setup jlmerclusterperm package","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"","code":"jlmerclusterperm_setup(..., restart = TRUE, verbose = TRUE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"... Ignored. restart Whether set fresh Julia session, given one already running. FALSE jlmerclusterperm_setup() already called, nothing happens. verbose Print progress messages Julia console","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"TRUE","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"","code":"# \\donttest{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":null,"dir":"Reference","previous_headings":"","what":"Tidier methods for Julia regression models — julia_model_tidiers","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"Tidier methods Julia regression models","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"","code":"# S3 method for jlmer_mod tidy(x, effects = c(\"var_model\", \"ran_pars\", \"fixed\"), ...) # S3 method for jlmer_mod glance(x, ...)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"x object class jlmer_mod effects One \"var_model\", \"ran_pars\", \"fixed\" ... Unused","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"data frame","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } # Fixed-effects only model mod1 <- to_jlmer(weight ~ 1 + Diet, ChickWeight) tidy(mod1) #> # A tibble: 4 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 103. 4.67 22.0 6.77e-107 #> 2 Diet2 20.0 7.87 2.54 1.11e- 2 #> 3 Diet3 40.3 7.87 5.12 3.01e- 7 #> 4 Diet4 32.6 7.91 4.12 3.73e- 5 glance(mod1) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 578 5 69.3 -3268. 6546. 6568. 2758693. 573 # Mixed model mod2 <- to_jlmer(weight ~ 1 + Diet + (1 | Chick), ChickWeight) tidy(mod2) #> # A tibble: 6 × 7 #> effect group term estimate std.error statistic p.value #> #> 1 fixed NA (Intercept) 102. 5.78 17.6 2.74e-69 #> 2 fixed NA Diet2 20.8 9.79 2.13 3.34e- 2 #> 3 fixed NA Diet3 41.2 9.79 4.20 2.63e- 5 #> 4 fixed NA Diet4 33.3 9.83 3.39 7.08e- 4 #> 5 ran_pars Chick sd__(Intercept) 15.7 NA NA NA #> 6 ran_pars Residual sd__Observation 67.3 NA NA NA glance(mod2) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 578 6 67.3 -3265. 6542. 6568. 6530. 572 # Select which of fixed/random effects to return tidy(mod2, effects = \"fixed\") #> # A tibble: 4 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 102. 5.78 17.6 2.74e-69 #> 2 Diet2 20.8 9.79 2.13 3.34e- 2 #> 3 Diet3 41.2 9.79 4.20 2.63e- 5 #> 4 Diet4 33.3 9.83 3.39 7.08e- 4 tidy(mod2, effects = \"ran_pars\") #> # A tibble: 2 × 3 #> group term estimate #> #> 1 Chick sd__(Intercept) 15.7 #> 2 Residual sd__Observation 67.3 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":null,"dir":"Reference","previous_headings":"","what":"Set/get options for Julia progress bar — julia_progress","title":"Set/get options for Julia progress bar — julia_progress","text":"Set/get options Julia progress bar","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Set/get options for Julia progress bar — julia_progress","text":"","code":"julia_progress(show, width)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Set/get options for Julia progress bar — julia_progress","text":"show Whether show progress bar. may also pass list \"show\" \"width\". width Width progress bar. \"auto\", adjusts progress bar width fit console.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Set/get options for Julia progress bar — julia_progress","text":"Previous values show width","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Set/get options for Julia progress bar — julia_progress","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) # } # Show current progress options julia_progress() #> $show #> [1] TRUE #> #> $width #> [1] 36 #> # Set options and save previous options old_progress_opts <- julia_progress(show = FALSE, width = 100) julia_progress() #> $show #> [1] FALSE #> #> $width #> [1] 100 #> # Restoring progress settings by passing a list of old options old_progress_opts #> $show #> [1] TRUE #> #> $width #> [1] 36 #> julia_progress(old_progress_opts) identical(julia_progress(), old_progress_opts) #> [1] TRUE # Alternatively, reset to default settings using this syntax: julia_progress(show = TRUE, width = \"auto\") # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":null,"dir":"Reference","previous_headings":"","what":"Interface to the Julia RNG — julia_rng","title":"Interface to the Julia RNG — julia_rng","text":"Interface Julia RNG","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Interface to the Julia RNG — julia_rng","text":"","code":"set_rng_state(i) reset_rng_state() get_rng_state() set_rng_seed(seed) get_rng_seed()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Interface to the Julia RNG — julia_rng","text":"Counter number seed Seed","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Interface to the Julia RNG — julia_rng","text":"current seed counter","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Interface to the Julia RNG — julia_rng","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } # RNG initializes to seed=1 counter=0 get_rng_seed() #> [1] 1 get_rng_state() #> [1] 0 # setter/getter for RNG counter set_rng_state(123) get_rng_state() #> [1] 123 # setter/getter for RNG seed set_rng_seed(2) get_rng_seed() #> [1] 2 # restore to initial setting (seed=1, counter=0) set_rng_seed(1) set_rng_state(0) # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":null,"dir":"Reference","previous_headings":"","what":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"Check Julia setup requirements jlmerclusterperm","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"","code":"julia_setup_ok()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"Boolean","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"","code":"julia_setup_ok() #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":null,"dir":"Reference","previous_headings":"","what":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"Create specifications object fitting regression models Julia","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"","code":"make_jlmer_spec( formula, data, subject = NULL, trial = NULL, time = NULL, drop_terms = NULL, ... )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"formula Model formula R syntax data data frame subject Column subjects data. trial Column trials data. Must uniquely identify time series within subject (example, column items counterbalanced design subject sees exactly one item). time Column time data. drop_terms (Optional) terms drop reconstructed model formula ... Unused, extensibility.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"object class jlmer_spec.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"","code":"# Bare specification object (minimal spec for fitting a global model) spec <- make_jlmer_spec(weight ~ 1 + Diet, ChickWeight) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Data: #> # A tibble: 578 × 4 #> weight Diet2 Diet3 Diet4 #> #> 1 42 0 0 0 #> 2 51 0 0 0 #> 3 59 0 0 0 #> # ℹ 575 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Constraints on specification for CPA: # 1) The combination of `subject`, `trial`, and `time` must uniquely identify rows in the data # 2) `time` must have constant sampling rate (i.e., evenly spaced values) spec_wrong <- make_jlmer_spec( weight ~ 1 + Diet, ChickWeight, time = \"Time\" ) #> ! Grouping column \"Time\" does not uniquely identify rows in the data #> ! Sampling rate for the `time` column \"Time\" is not constant - may affect interpretability of results. unique(ChickWeight$Time) #> [1] 0 2 4 6 8 10 12 14 16 18 20 21 # Corrected specification for the above spec_correct <- make_jlmer_spec( weight ~ 1 + Diet, subset(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec_correct #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":null,"dir":"Reference","previous_headings":"","what":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"Permute data respecting grouping structure(s) observations","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"","code":"permute_by_predictor( jlmer_spec, predictors, predictor_type = c(\"guess\", \"between_participant\", \"within_participant\"), n = 1L )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() predictors vector terms model. multiple, must form levels one predictor. predictor_type Whether predictor \"between_participant\" \"within_participant\". Defaults \"guess\". n Number permuted samples data generate. Defaults 1L.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"long dataframe permuted re-samples .id column representing replication IDs.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } # Example data setup chickweights_df <- ChickWeight chickweights_df <- chickweights_df[chickweights_df$Time <= 20, ] chickweights_df$DietInt <- as.integer(chickweights_df$Diet) head(chickweights_df) #> Grouped Data: weight ~ Time | Chick #> weight Time Chick Diet DietInt #> 1 42 0 1 1 1 #> 2 51 2 1 1 1 #> 3 59 4 1 1 1 #> 4 64 6 1 1 1 #> 5 76 8 1 1 1 #> 6 93 10 1 1 1 # Example 1: Spec object using the continuous `DietInt` predictor chickweights_spec1 <- make_jlmer_spec( formula = weight ~ 1 + DietInt, data = chickweights_df, subject = \"Chick\", time = \"Time\" ) chickweights_spec1 #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + DietInt #> Predictors: #> DietInt: DietInt #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight DietInt Chick Time #> #> 1 42 1 1 0 #> 2 51 1 1 2 #> 3 59 1 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Shuffling `DietInt` values guesses `predictor_type = \"between_participant\"` reset_rng_state() spec1_perm1 <- permute_by_predictor(chickweights_spec1, predictors = \"DietInt\") #> ℹ Guessed `predictor_type` to be \"between_participant\" # This calls the same shuffling algorithm for CPA in Julia, so counter is incremented get_rng_state() #> [1] 38 # Shuffling under shared RNG state reproduces the same permutation of the data reset_rng_state() spec1_perm2 <- permute_by_predictor(chickweights_spec1, predictors = \"DietInt\") #> ℹ Guessed `predictor_type` to be \"between_participant\" identical(spec1_perm1, spec1_perm2) #> [1] TRUE # Example 2: Spec object using the multilevel `Diet` predictor chickweights_spec2 <- make_jlmer_spec( formula = weight ~ 1 + Diet, data = chickweights_df, subject = \"Chick\", time = \"Time\" ) chickweights_spec2 #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Levels of a category are automatically shuffled together reset_rng_state() spec2_perm1 <- permute_by_predictor(chickweights_spec2, predictors = \"Diet2\") #> ℹ Guessed `predictor_type` to be \"between_participant\" #> ℹ Shuffling all levels of the factor together (\"Diet2\", \"Diet3\", and \"Diet4\") reset_rng_state() spec2_perm2 <- permute_by_predictor(chickweights_spec2, predictors = c(\"Diet2\", \"Diet3\", \"Diet4\")) #> ℹ Guessed `predictor_type` to be \"between_participant\" identical(spec2_perm1, spec2_perm2) #> [1] TRUE # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":null,"dir":"Reference","previous_headings":"","what":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"Simulate cluster-mass statistics via bootstrapped permutations","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"","code":"permute_timewise_statistics( jlmer_spec, family = c(\"gaussian\", \"binomial\"), statistic = c(\"t\", \"chisq\"), nsim = 100L, predictors = NULL, ... )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". statistic Test statistic calculating cluster mass. Can one \"t\" (default) regression model output \"chisq\" likelihood ratio test (takes twice long calculate). nsim Number simulations description predictors (Optional) subset predictors test. Defaults NULL tests predictors. ... Optional arguments passed Julia model fitting. Defaults fast = TRUE (family = \"binomial\") progress = FALSE.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"simulation--time--predictor 3D array cluster statistics, class timewise_statistics.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Simulation x Time x Predictor array of t-statistics from regression output reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 3) round(null_statistics, 2) #> , , Predictor = Diet2 #> #> Time #> Sim 0 2 4 6 8 10 12 14 16 18 20 #> 1 -0.90 0.41 0.39 1.12 1.58 2.13 2.34 2.58 2.47 2.67 2.61 #> 2 1.40 0.90 0.96 1.44 1.61 1.45 1.23 0.90 0.80 0.78 0.76 #> 3 -1.44 1.23 0.73 0.88 0.93 0.69 0.58 0.61 0.77 1.04 1.43 #> #> , , Predictor = Diet3 #> #> Time #> Sim 0 2 4 6 8 10 12 14 16 18 20 #> 1 0.00 -0.21 0.11 -0.68 -0.66 -0.38 -0.01 -0.02 -0.17 -0.20 -0.12 #> 2 0.70 0.56 -0.10 0.17 0.37 0.37 0.55 0.21 0.46 0.44 0.02 #> 3 0.72 1.90 0.85 0.20 0.19 -0.09 -0.34 -0.24 0.34 0.44 0.83 #> #> , , Predictor = Diet4 #> #> Time #> Sim 0 2 4 6 8 10 12 14 16 18 20 #> 1 0.45 -0.82 -0.33 -0.68 -0.36 -0.35 -0.41 0.19 -0.04 -0.16 -0.16 #> 2 2.09 -0.69 -0.10 0.34 0.56 0.63 0.35 0.05 -0.05 -0.26 -0.43 #> 3 -2.16 -1.38 -1.35 -0.85 -0.02 -0.12 0.01 0.59 0.71 0.67 0.53 #> # Collect as dataframe with `tidy()` permuted_timewise_stats_df <- tidy(null_statistics) permuted_timewise_stats_df #> # A tibble: 99 × 4 #> predictor time statistic sim #> #> 1 Diet2 0 -8.98e- 1 1 #> 2 Diet2 0 1.40e+ 0 2 #> 3 Diet2 0 -1.44e+ 0 3 #> 4 Diet3 0 2.34e-15 1 #> 5 Diet3 0 6.98e- 1 2 #> 6 Diet3 0 7.21e- 1 3 #> 7 Diet4 0 4.49e- 1 1 #> 8 Diet4 0 2.09e+ 0 2 #> 9 Diet4 0 -2.16e+ 0 3 #> 10 Diet2 2 4.12e- 1 1 #> # ℹ 89 more rows # Permutations ran under the same RNG state are identical reset_rng_state() null_statistics2 <- permute_timewise_statistics(spec, nsim = 3) identical(null_statistics, null_statistics2) #> [1] TRUE get_rng_state() #> [1] 111 null_statistics3 <- permute_timewise_statistics(spec, nsim = 3) identical(null_statistics, null_statistics3) #> [1] FALSE # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/reexports.html","id":null,"dir":"Reference","previous_headings":"","what":"Objects exported from other packages — reexports","title":"Objects exported from other packages — reexports","text":"objects imported packages. Follow links see documentation. generics glance, tidy","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit a Julia regression model using lme4 syntax — to_jlmer","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"Fit Julia regression model using lme4 syntax","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"","code":"to_jlmer( formula, data, family = c(\"gaussian\", \"binomial\"), jlmer_spec_opts = list(), ..., progress = FALSE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"formula Model formula R syntax data data frame family GLM family. Currently supports \"gaussian\" \"binomial\". jlmer_spec_opts List options passed make_jlmer_spec() ... Optional arguments passed Julia model fitting. progress TRUE, prints timing iterations.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"jlmer_mod object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } jlmerclusterperm_setup(verbose = FALSE) # Fitting a regression model with R formula syntax to_jlmer(weight ~ 1 + Diet, ChickWeight) #> #> ──────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ──────────────────────────────────────────────────────────────────────── #> (Intercept) 102.645 4.67395 21.96 <1e-99 93.4847 111.806 #> Diet2 19.9712 7.86744 2.54 0.0111 4.55132 35.3911 #> Diet3 40.3045 7.86744 5.12 <1e-06 24.8847 55.7244 #> Diet4 32.6173 7.91046 4.12 <1e-04 17.113 48.1215 #> ──────────────────────────────────────────────────────────────────────── # `lm()` equivalent summary(lm(weight ~ 1 + Diet, ChickWeight))$coef #> Estimate Std. Error t value Pr(>|t|) #> (Intercept) 102.64545 4.673954 21.961161 4.712762e-78 #> Diet2 19.97121 7.867437 2.538465 1.139708e-02 #> Diet3 40.30455 7.867437 5.122958 4.113938e-07 #> Diet4 32.61726 7.910461 4.123307 4.286352e-05 # Fitting a mixed model with {lme4} syntax to_jlmer(weight ~ 1 + Diet + (1 | Chick), ChickWeight) #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 247.2559 15.7244 #> Residual 4527.9661 67.2902 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 101.781 5.78499 17.59 <1e-68 #> Diet2 20.8353 9.79412 2.13 0.0334 #> Diet3 41.1687 9.79412 4.20 <1e-04 #> Diet4 33.284 9.82881 3.39 0.0007 #> ────────────────────────────────────────────────── # Passing MixedModels.jl fit options to the `...` to_jlmer(weight ~ 1 + Diet + (1 | Chick), ChickWeight, REML = TRUE) #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 305.7755 17.4864 #> Residual 4526.5647 67.2797 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 101.634 6.04044 16.83 <1e-62 #> Diet2 20.9832 10.2365 2.05 0.0404 #> Diet3 41.3165 10.2365 4.04 <1e-04 #> Diet4 33.4041 10.27 3.25 0.0011 #> ────────────────────────────────────────────────── # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":null,"dir":"Reference","previous_headings":"","what":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"Test probability cluster-mass statistics range threshold values","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"","code":"walk_threshold_steps( empirical_statistics, null_statistics, steps, top_n = Inf, binned = FALSE, add1 = TRUE, progress = TRUE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"empirical_statistics predictor--time matrix empirical timewise statistics. null_statistics simulation--time--predictor 3D array null (permuted) timewise statistics. steps vector threshold values test top_n many clusters return, order size cluster-mass statistic. Defaults Inf return detected clusters. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist. add1 Whether add 1 numerator denominator calculating p-value. Use TRUE effectively count observed statistic part permuted null distribution (recommended larger nsim prior publishing results). progress Whether display progress bar","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"data frame predictor clusters-mass statistics threshold.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Compute timewise statistics for the observed and permuted data empirical_statistics <- compute_timewise_statistics(spec) reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) # Test cluster mass/probability under different threshold values walk_threshold_steps(empirical_statistics, null_statistics, steps = 1:3, progress = FALSE) #> # A tibble: 9 × 8 #> threshold predictor id start end length sum_statistic pvalue #> #> 1 1 Diet2 1 2 20 10 19.4 0.00990 #> 2 1 Diet3 1 2 20 10 34.6 0.00990 #> 3 1 Diet4 1 2 20 10 39.4 0.00990 #> 4 2 Diet2 1 4 8 3 8.50 0.0297 #> 5 2 Diet3 1 2 20 10 34.6 0.00990 #> 6 2 Diet4 1 2 20 10 39.4 0.00990 #> 7 3 Diet3 1 4 8 3 12.7 0.00990 #> 8 3 Diet3 2 16 20 3 10.4 0.00990 #> 9 3 Diet4 1 2 12 6 29.7 0.00990 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-104","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.4","title":"jlmerclusterperm 1.0.4","text":"user-visible changes Restructured internal Julia code module (JlmerClusterPerm.jl)","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-103","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.3","title":"jlmerclusterperm 1.0.3","text":"CRAN release: 2023-07-19 Fixes CRAN: Use string package version comparison","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-102","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.2","title":"jlmerclusterperm 1.0.2","text":"CRAN release: 2023-07-15","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"bug-fixes-1-0-2","dir":"Changelog","previous_headings":"","what":"Bug fixes","title":"jlmerclusterperm 1.0.2","text":"jlmerclusterperm_setup() now works rc alpha builds Julia Fixes CRAN: exclusion Manifest.toml now specified .Rbuildignore - Project.toml bundled intended.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-101","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.1","title":"jlmerclusterperm 1.0.1","text":"CRAN release: 2023-07-02 jlmerclusterperm_setup() now exits early Julia version requirement (>=1.8) met. Fixes CRAN: Check Julia version requirement proceeding examples tests","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-100","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.0","title":"jlmerclusterperm 1.0.0","text":"CRAN release: 2023-06-27","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"breaking-changes-1-0-0","dir":"Changelog","previous_headings":"","what":"Breaking changes","title":"jlmerclusterperm 1.0.0","text":"threshold_steps argument walk_threshold_steps() renamed steps.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"new-features-1-0-0","dir":"Changelog","previous_headings":"","what":"New features","title":"jlmerclusterperm 1.0.0","text":"New functions interface Julia RNG seed: get_rng_seed() set_rng_seed()","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"other-improvements-1-0-0","dir":"Changelog","previous_headings":"","what":"Other improvements","title":"jlmerclusterperm 1.0.0","text":"jlmerclusterperm_setup() now echos Pkg.instantiate() print precompilation information upon first setup call","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-020","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 0.2.0","title":"jlmerclusterperm 0.2.0","text":"Added vignettes. Significant usability improvements","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-010","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 0.1.0","title":"jlmerclusterperm 0.1.0","text":"Initial release","code":""}] +[{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Asynchronous CPA","text":"Minimal example data CPA specification: Example CPA (1000 simulations): details actual CPA skipped ’s focus vignette, example readme just nsim.","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) chickweights <- ChickWeight chickweights$Time <- as.integer(factor(chickweights$Time)) chickweights_spec <- make_jlmer_spec( formula = weight ~ 1 + Diet, data = chickweights, subject = \"Chick\", time = \"Time\" ) set_rng_state(123L) CPA <- clusterpermute( chickweights_spec, threshold = 2.5, nsim = 1000, progress = FALSE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"basic-idea","dir":"Articles","previous_headings":"","what":"Basic idea","title":"Asynchronous CPA","text":"basic idea behind asynchronous CPA strategy start background R process whose sole job send instructions CPA Julia. ’s asynchronous running CPA way block interactive R session. Since work done Julia anyways, parallelization virtually impact performance evaluating R code. future package allows asynchronous evaluation R code. ’s big package implementing complex topic. can read project futureverse just show bare minimum . advanced users, note ’re use parallelization asynchronous properties (non-blocking evaluation R code background process). recommended start multiple R processes running CPA Julia session shared already multithreaded (can spare cores, set options(\"jlmerclusterperm.nthreads\") calling jlmerclusterperm_setup()). Later updates jlmerclusterperm may wrap workflow principled way, now vignette serves minimally working example asynchronously running CPA.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"setup-for-async-cpa","dir":"Articles","previous_headings":"Walkthrough","what":"Setup for async CPA","title":"Asynchronous CPA","text":"Three things order workflow: Load future package Initialize multisession future Grab options jlmerclusterperm package environment Please treat .jlmerclusterperm internal variable read-object - ’s unexported meant manipulated.","code":"library(future) plan(multisession) pkgopts <- as.list(jlmerclusterperm:::.jlmerclusterperm)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"creating-the-future-object","dir":"Articles","previous_headings":"Walkthrough","what":"Creating the future object","title":"Asynchronous CPA","text":"start creating special object class using future::future(). can simply pass clusterpermute() code first argument future(), slight modification ensure background process connects Julia session. template follows. First, make pkgopts object (defined ) available future (via globals argument). , inside future expression ensure background process shares Julia session (via list2env(...)). future object replicating 1000-simulation CPA look like following: future created, immediately starts executing code background process. important evaluate future object (f) directly. Instead, query future::resolved() - simply tells whether background evaluation completed : , use loop show background CPA non-blocking. evaluating R code interactive session simultaneously CPA running: reach end loop 4.5 seconds, approximately long CPA took complete (confirm next section). point future completed result available collection. Although R code just crude progress alert using Sys.sleep(), can freely evaluate R code except functions call Julia. can make plots, clean data, write analyses, etc., just don’t run another CPA one already running Julia process shared.","code":"# Not run future( { list2env(pkgopts, jlmerclusterperm:::.jlmerclusterperm) ## Your CPA code below ## }, globals = structure(TRUE, add = \"pkgopts\") ) f <- future( { list2env(as.list(pkgopts), jlmerclusterperm:::.jlmerclusterperm) ## Your CPA code below ## set_rng_state(123L) clusterpermute( chickweights_spec, threshold = 2.5, nsim = 1000 ) }, globals = structure(TRUE, add = \"pkgopts\") ) resolved(f) #> [1] FALSE i <- 0 while (!resolved(f)) { Sys.sleep(0.5) i <<- i + 1 cat(sprintf(\"Elapsed: %.01fs\", i / 2), \"\\n\") } #> Elapsed: 0.5s #> Elapsed: 1.0s #> Elapsed: 1.5s #> Elapsed: 2.0s #> Elapsed: 2.5s #> Elapsed: 3.0s #> Elapsed: 3.5s #> Elapsed: 4.0s #> Elapsed: 4.5s resolved(f) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/asynchronous-cpa.html","id":"collecting-the-results","dir":"Articles","previous_headings":"Walkthrough","what":"Collecting the results","title":"Asynchronous CPA","text":"can collect output background process future::value(). first, double check make sure background process indeed finished evaluating: ’s confirmed, can use value() collect results future assign variable inspection: Note value() also prints messages encountered executing CPA code. see specify progress = FALSE clusterpermute() call passed future. purposes, just note adding times messages similar saw loop (4.5 seconds). output asynchronous CPA (CPA_async) identical initial CPA ran beginning (CPA) ran default seed 1 RNG counter value 123L: separate vignette covers Julia RNG. practice, may want use random seed (via set_rng_seed()) background CPA.","code":"resolved(f) #> [1] TRUE CPA_async <- value(f) #> Connecting to Julia TCP server at localhost:11981 ... #> ℹ Detecting empirical clusters and calculating cluster-mass statistics. #> ✔ Detecting empirical clusters and calculating cluster-mass statistics. [125ms] #> #> ℹ Sampling cluster-mass statistics from a bootstrapped null distribution. #> ✔ Sampling cluster-mass statistics from a bootstrapped null distribution. [4.8s] #> #> ℹ Calculating the probability of the observed cluster-mass statistics. #> ✔ Calculating the probability of the observed cluster-mass statistics. [28ms] #> identical(CPA_async, CPA) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"background","dir":"Articles","previous_headings":"","what":"Background","title":"de Carvalho et al. 2021","text":"data comes eye-tracking study de Carvalho, Dautriche, Fiévet, & Christophe (2021) “Toddlers exploit referential syntactic cues flexibly adapt interpretation novel verb meanings.” article reproduces expands Experiment 2 analysis, used CPA (t-tests) conduct pairwise comparison three conditions (relevant Figure 7 paper ). data comes mostly ready CPA straight study’s OSF repository. collapsible chunk contains code prepare data reproduce figure. Data cleaning Code Figure 7 plots data original study (close reproduce ). E2_data_agg dataframe following four columns: Subject: Unique identifier subjects Condition: -participant factor variable three levels (\"Intransitive\", \"RightDislocated\", \"Transitive\") Time: continuous measure time 0-8000ms 50ms intervals Prop: Proportion looks target (averaged across trials within condition, subject) following reproduction Figure 7 original paper: original study used pairwise comparisons analyze relationship three conditions. smaller figures (B, C, D) plot following relationships: (B) looks target \"Transitive\" compared \"RightDislocated\" (C) looks target \"RightDislocated\" compared \"Intransitive\" (D) looks target \"Transitive\" compared \"Intransitive\" rest vignette organized follows: First, replicate individual pairwise CPAs original paper. Next, consider parsimonious analysis reformulates research hypothesis choice regression contrast. Lastly, build model (2) consider issues around model diagnostics complexity (linear vs. logistic; fixed vs. mixed) context CPA Load package start Julia instance jlmerclusterperm_setup() proceeding.","code":"library(dplyr) library(forcats) filepath <- \"https://osf.io/download/var2h/?view_only=2a06746ba360446e9857df73307d21be\" E2_data_raw <- readr::read_delim(filepath) E2_data <- E2_data_raw %>% filter(away != 1) %>% mutate(Target = as.integer(bin_2P == 1)) %>% mutate(Condition = fct_recode( factor(Condition, levels = c(\"intrans\", \"pros\", \"trans\")), \"Intransitive\" = \"intrans\", \"RightDislocated\" = \"pros\", \"Transitive\" = \"trans\" )) %>% select(Subject, Trial, Condition, Time, Target) E2_data_agg <- E2_data %>% group_by(Subject, Condition, Time) %>% summarize(Prop = mean(Target), .groups = \"drop\") # Unaggregated trial-level data of 1s and 0s E2_data #> # A tibble: 69,123 × 5 #> Subject Trial Condition Time Target #> #> 1 200.asc 0 RightDislocated 0 0 #> 2 200.asc 0 RightDislocated 400 0 #> 3 200.asc 0 RightDislocated 450 1 #> 4 200.asc 0 RightDislocated 500 1 #> 5 200.asc 0 RightDislocated 550 1 #> 6 200.asc 0 RightDislocated 600 1 #> 7 200.asc 0 RightDislocated 650 1 #> 8 200.asc 0 RightDislocated 700 1 #> 9 200.asc 0 RightDislocated 750 1 #> 10 200.asc 0 RightDislocated 800 1 #> # ℹ 69,113 more rows # Aggregated subject-mean proportions data used in original study E2_data_agg #> # A tibble: 11,540 × 4 #> Subject Condition Time Prop #> #> 1 200.asc RightDislocated 0 0.5 #> 2 200.asc RightDislocated 50 1 #> 3 200.asc RightDislocated 100 1 #> 4 200.asc RightDislocated 150 1 #> 5 200.asc RightDislocated 200 1 #> 6 200.asc RightDislocated 250 0.5 #> 7 200.asc RightDislocated 300 0.5 #> 8 200.asc RightDislocated 350 0.4 #> 9 200.asc RightDislocated 400 0.286 #> 10 200.asc RightDislocated 450 0.429 #> # ℹ 11,530 more rows library(ggplot2) #> Warning: package 'ggplot2' was built under R version 4.3.1 make_fig7_plot <- function(conditions) { E2_data %>% group_by(Condition, Time) %>% summarize( Prop = mean(Target), se = sqrt(var(Target) / n()), lower = Prop - se, upper = Prop + se, .groups = \"drop\" ) %>% filter(Condition %in% conditions) %>% ggplot(aes(Time, Prop, color = Condition, fill = Condition)) + geom_hline(aes(yintercept = .5)) + geom_ribbon( aes(ymin = lower, ymax = upper), alpha = .2, show.legend = FALSE, ) + geom_line(linewidth = 1) + scale_color_manual( aesthetics = c(\"color\", \"fill\"), values = setNames(scales::hue_pal()(3)[c(2, 1, 3)], levels(E2_data$Condition)) ) + scale_y_continuous(limits = c(.2, .8), oob = scales::oob_keep) + labs(y = NULL, x = NULL) + theme_minimal() + theme(axis.title.y = element_text(angle = 0, vjust = .5, hjust = 0)) } fig7_comparisons <- list( \"A\" = c(\"Intransitive\", \"RightDislocated\", \"Transitive\"), \"B\" = c(\"RightDislocated\", \"Transitive\"), \"C\" = c(\"Intransitive\", \"RightDislocated\"), \"D\" = c(\"Intransitive\", \"Transitive\") ) fig7 <- lapply(fig7_comparisons, make_fig7_plot) # Figure 7 combined plots library(patchwork) #> Warning: package 'patchwork' was built under R version 4.3.1 p_top <- fig7$A + scale_x_continuous(breaks = scales::breaks_width(1000)) + theme(legend.position = \"bottom\") p_bot <- (fig7$B + fig7$C + fig7$D) & guides(color = guide_none()) fig7_combined <- p_top / guide_area() / p_bot + plot_layout(guides = \"collect\", heights = c(2, .1, 1)) + plot_annotation(tag_levels = \"A\") E2_data_agg #> # A tibble: 11,540 × 4 #> Subject Condition Time Prop #> #> 1 200.asc RightDislocated 0 0.5 #> 2 200.asc RightDislocated 50 1 #> 3 200.asc RightDislocated 100 1 #> 4 200.asc RightDislocated 150 1 #> 5 200.asc RightDislocated 200 1 #> 6 200.asc RightDislocated 250 0.5 #> 7 200.asc RightDislocated 300 0.5 #> 8 200.asc RightDislocated 350 0.4 #> 9 200.asc RightDislocated 400 0.286 #> 10 200.asc RightDislocated 450 0.429 #> # ℹ 11,530 more rows fig7_combined library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"replicating-the-pairwise-cpas","dir":"Articles","previous_headings":"","what":"Replicating the pairwise CPAs","title":"de Carvalho et al. 2021","text":"three conditions experiment levels Condition factor variable: begin replication \"Transitive\" vs. \"RightDislocated\" comparison shown Figure 7B apply logic two pairwise comparisons 7C 7D.","code":"levels(E2_data_agg$Condition) #> [1] \"Intransitive\" \"RightDislocated\" \"Transitive\""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"transitive-vs--rightdislocated","dir":"Articles","previous_headings":"Replicating the pairwise CPAs","what":"Transitive vs. RightDislocated","title":"de Carvalho et al. 2021","text":"reproduced Figure 7B compares \"Transitive\" \"RightDislocated\" conditions. paper (caption Figure 7; emphasis mine) reports: transitive right-dislocated conditions differed second repetition novel verbs (~6400 ms onset test trials end trials). now replicate analysis. First, prepare specification object. Two things note : express original t-test regression model Condition predictor drop third, unused condition data factor representation Next, fit global model sanity check structure model output. get one estimate ConditionTransitive positive coefficient, ’d expect: Finally, call clusterpermute() threshold = 1.5 (original study) simulate 100 permutations: detect largest empirical cluster spanning 6400-8000ms reported original paper. converges around p=0.02 separate 10,000-simulation run (shown ).","code":"fig7$B spec_7B <- make_jlmer_spec( formula = Prop ~ Condition, data = E2_data_agg %>% filter(Condition %in% c(\"Transitive\", \"RightDislocated\")) %>% mutate(Condition = droplevels(Condition)), # or forcats::fct_drop() subject = \"Subject\", time = \"Time\" ) spec_7B #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: Prop ~ 1 + ConditionTransitive #> Predictors: #> Condition: ConditionTransitive #> Groupings: #> Subject: Subject #> Trial: #> Time: Time #> Data: #> # A tibble: 7,688 × 4 #> Prop ConditionTransitive Subject Time #> #> 1 0.5 0 200.asc 0 #> 2 1 0 200.asc 50 #> 3 1 0 200.asc 100 #> # ℹ 7,685 more rows #> ──────────────────────────────────────────────────────────────────────────────── jlmer(spec_7B) #> #> ────────────────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ────────────────────────────────────────────────────────────────────────────────── #> (Intercept) 0.592999 0.00364424 162.72 <1e-99 0.585856 0.600141 #> ConditionTransitive 0.0569017 0.0051591 11.03 <1e-27 0.04679 0.0670133 #> ────────────────────────────────────────────────────────────────────────────────── clusterpermute(spec_7B, threshold = 1.5, nsim = 100L, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionTransitive (n = 100) #> Mean (SD): -2.674 (27.83) #> Coverage intervals: 95% [-61.740, 46.439] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionTransitive #> [300, 1150]: 40.643 (p=0.1287) #> [2700, 2750]: -4.068 (p=0.9010) #> [6150, 6200]: 3.249 (p=0.9505) #> [6400, 8000]: 88.742 (p=0.0198) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"rightdislocated-vs--intransitive","dir":"Articles","previous_headings":"Replicating the pairwise CPAs","what":"RightDislocated vs. Intransitive","title":"de Carvalho et al. 2021","text":"reproduced Figure 7C compares \"RightDislocated\" \"Intransitive\" conditions. paper reports: intransitive right-dislocated conditions differed first repetition novel verbs (2100 ms 3500 ms beginning test trials). repeat CPA procedure pairwise comparison: largest empirical cluster detect spans 2150-3650ms, slightly different cluster reported original paper (2100-3500ms). relatively less “extreme” cluster converges around p=0.05 10,000-simulation run.","code":"fig7$C spec_7C <- make_jlmer_spec( formula = Prop ~ Condition, data = E2_data_agg %>% filter(Condition %in% c(\"RightDislocated\", \"Intransitive\")) %>% mutate(Condition = droplevels(Condition)), subject = \"Subject\", time = \"Time\" ) clusterpermute(spec_7C, threshold = 1.5, nsim = 100L, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionRightDislocated (n = 100) #> Mean (SD): 0.619 (33.26) #> Coverage intervals: 95% [-63.734, 67.326] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionRightDislocated #> [150, 200]: 3.562 (p=0.9010) #> [2150, 3650]: 72.842 (p=0.0297) #> [4550, 5050]: 23.836 (p=0.3960) #> [5150, 5350]: 8.589 (p=0.7723) #> [5700, 5750]: 3.332 (p=0.9208) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"transitive-vs--intransitive","dir":"Articles","previous_headings":"Replicating the pairwise CPAs","what":"Transitive vs. Intransitive","title":"de Carvalho et al. 2021","text":"reproduced Figure 7D compares \"Transitive\" \"Intransitive\" conditions. paper reports: transitive intransitive conditions differed slightly offset first sentence test trials (4500 ms beginning test trials end trials). repeat CPA procedure pairwise comparison: largest empirical cluster detect spans 4600-8000ms, slightly different cluster reported original paper (4500-8000ms). converges around p=0.001 separate 10,000-simulation run.","code":"fig7$D spec_7D <- make_jlmer_spec( formula = Prop ~ Condition, data = E2_data_agg %>% filter(Condition %in% c(\"Transitive\", \"Intransitive\")) %>% mutate(Condition = droplevels(Condition)), subject = \"Subject\", time = \"Time\" ) clusterpermute(spec_7D, threshold = 1.5, nsim = 100L, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionTransitive (n = 100) #> Mean (SD): 2.024 (41.73) #> Coverage intervals: 95% [-70.510, 88.501] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionTransitive #> [300, 1000]: 25.928 (p=0.4356) #> [1300, 1400]: 4.808 (p=0.8713) #> [2800, 2900]: 5.356 (p=0.8218) #> [3000, 4300]: 52.638 (p=0.2079) #> [4600, 8000]: 172.635 (p=0.0198) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"expressed-as-regression-contrasts","dir":"Articles","previous_headings":"","what":"Expressed as regression contrasts","title":"de Carvalho et al. 2021","text":"now consider parsimonious analysis translates research hypothesis contrast coding avoid multiple testing. Specifically, exploit fact original paper specifies hypothesis Intransitive < RightDislocated < Transitive.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"helmert-deviation-coding","dir":"Articles","previous_headings":"Expressed as regression contrasts","what":"Helmert (deviation) coding","title":"de Carvalho et al. 2021","text":"Testing ordinal relationship levels category require possible pairwise comparisons; instead, can approximated via Helmert coding (.k.. deviance coding) K levels expressed K-1 contrasts, contrast successively comparing level vs. average previous (typically lower) levels. Critically, Helmert contrasts orthogonal, can test simultaneously single model. data, test ordinal relationship Intransitive < RightDislocated < Transitive via two contrasts: \"RightDislocated\" vs. \"Intransitive\" \"Transitive\" vs. average \"RightDislocated\" \"Intransitive\" corresponding numerical coding following: practice, Helmert contrasts often standardized deviations expressed unit 1. also comparison \"RightDislocated\" vs. \"Intransitive\" expressed 1 unit RvsI comparison \"Transitive\" vs. average \"RightDislocated\" \"Intransitive\" expressed 1 unit TvsRI: contrast matrix, make new column original data called ConditionHelm copying Condition column, apply contrasts new column: Lastly, build new specification making use full data. , predict Prop ConditionHelm estimate effect contrasts single model. sanity check, fit global model - expect estimate contrast indeed find RvsI TvsRI output positive coefficients. suggests ordinal relationship three conditions hold least globally. Note coefficient RvsI contrast exactly pairwise model using spec_7C earlier, also compared \"RightDislocated\" \"Intransitive\": spec_helm two conditions RvsI coded -0.5 0.5, spec_7C’s treatment coding coded 0 1. Since express difference \"Intransitive\" \"RightDislocated\" unit +1, two coefficients equal magnitude sign.","code":"condition_helm <- contr.helmert(3) colnames(condition_helm) <- c(\"RvsI\", \"TvsRI\") condition_helm #> RvsI TvsRI #> 1 -1 -1 #> 2 1 -1 #> 3 0 2 condition_helm[, 1] <- condition_helm[, 1] / 2 condition_helm[, 2] <- condition_helm[, 2] / 3 condition_helm #> RvsI TvsRI #> 1 -0.5 -0.3333333 #> 2 0.5 -0.3333333 #> 3 0.0 0.6666667 E2_data_agg$ConditionHelm <- E2_data_agg$Condition contrasts(E2_data_agg$ConditionHelm) <- condition_helm # For pretty-printing as fractions MASS::fractions(contrasts(E2_data_agg$ConditionHelm)) #> RvsI TvsRI #> Intransitive -1/2 -1/3 #> RightDislocated 1/2 -1/3 #> Transitive 0 2/3 spec_helm <- make_jlmer_spec( formula = Prop ~ ConditionHelm, data = E2_data_agg, subject = \"Subject\", time = \"Time\" ) spec_helm #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: Prop ~ 1 + ConditionHelmRvsI + ConditionHelmTvsRI #> Predictors: #> ConditionHelm: ConditionHelmRvsI, ConditionHelmTvsRI #> Groupings: #> Subject: Subject #> Trial: #> Time: Time #> Data: #> # A tibble: 11,540 × 5 #> Prop ConditionHelmRvsI ConditionHelmTvsRI Subject Time #> #> 1 0.5 0.5 -0.333 200.asc 0 #> 2 1 0.5 -0.333 200.asc 50 #> 3 1 0.5 -0.333 200.asc 100 #> # ℹ 11,537 more rows #> ──────────────────────────────────────────────────────────────────────────────── jlmer(spec_helm) %>% tidy(effects = \"fixed\") #> # A tibble: 3 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.591 0.00215 275. 0 #> 2 ConditionHelmRvsI 0.0640 0.00526 12.2 5.67e-34 #> 3 ConditionHelmTvsRI 0.0889 0.00457 19.5 2.02e-84 jlmer(spec_7C) %>% tidy(effects = \"fixed\") #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.529 0.00379 140. 0 #> 2 ConditionRightDislocated 0.0640 0.00536 11.9 8.30e-33"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"interpreting-cpa-results","dir":"Articles","previous_headings":"Expressed as regression contrasts","what":"Interpreting CPA results","title":"de Carvalho et al. 2021","text":"proceed normal clusterpermute() using new spec_helm: ’s summary find CPA_helm: RvsI clusters similar detected previous “partial” CPA spec_7C. slight differences cluster-mass due part due Helmert-coded model simultaneously estimating TvsRI contrast. importantly, CPA_helm detects largest cluster \"RightDislocated\" \"Intransitive\" (2150-3650ms). converges around p=0.05 10,000-simulation run. TvsRI clusters new, largest cluster predictor spans 5800ms-8000ms. converges around p=0.01 separate 10,000-simulation run. cluster effectively captures region relationship Transitive > (RightDislocated & Intransitive) emerges robust. Essentially, TvsRI comparison line \"Transitive\" invisible line runs lines \"Intransitive\" \"RightDislocated\". conclude visualizing clusters two Helmert-coded terms, annotated empirical data. “invisible” line RI Helmert coding drawn dashed line.","code":"reset_rng_state() CPA_helm <- clusterpermute(spec_helm, threshold = 1.5, nsim = 100L, progress = FALSE) CPA_helm #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionHelmRvsI (n = 100) #> Mean (SD): 4.019 (28.80) #> Coverage intervals: 95% [-45.805, 67.827] #> ConditionHelmTvsRI (n = 100) #> Mean (SD): 1.213 (24.59) #> Coverage intervals: 95% [-50.218, 47.462] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionHelmRvsI #> [150, 200]: 3.654 (p=0.8911) #> [2150, 3650]: 71.141 (p=0.0297) #> [4550, 5350]: 36.043 (p=0.2079) #> [5700, 5750]: 3.328 (p=0.9010) #> ConditionHelmTvsRI #> [300, 1050]: 35.863 (p=0.1881) #> [3500, 3700]: 8.348 (p=0.6436) #> [3850, 4250]: 15.985 (p=0.4455) #> [4800, 5600]: 36.143 (p=0.1782) #> [5800, 8000]: 123.631 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── fig7A_v2 <- fig7$A + geom_line( aes(Time, Prop, linetype = \"RI\"), inherit.aes = FALSE, data = . %>% filter(Condition %in% c(\"RightDislocated\", \"Intransitive\")) %>% group_by(Time) %>% summarize(Prop = mean(Prop)), ) + scale_linetype_manual(values = \"41\", guide = guide_legend(\"ConditionHelm\")) + guides(x = guide_none(\"\")) clusters_annotation <- tidy(CPA_helm$empirical_clusters) %>% mutate(contrast = gsub(\".*([TR])vs([RI]+)\", \"\\\\1 vs. \\\\2\", predictor)) %>% ggplot(aes(y = fct_rev(contrast))) + geom_segment( aes( x = start, xend = end, yend = contrast, color = pvalue < 0.05 ), linewidth = 8 ) + scale_color_manual(values = c(\"grey70\", \"steelblue\")) + scale_y_discrete() + scale_x_continuous(n.breaks = 9, limits = range(E2_data_agg$Time)) + theme_minimal() + theme( axis.title = element_blank(), axis.text.y = element_text(face = \"bold\"), panel.border = element_rect(fill = NA), panel.grid.major.y = element_blank() ) fig7A_v2 / clusters_annotation & plot_layout(heights = c(4, 1))"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"model-complexity","dir":"Articles","previous_headings":"","what":"Model complexity","title":"de Carvalho et al. 2021","text":"wrap case study considering complex CPA uses logistic mixed effects models trial-level data fixations target (1s 0s).","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"logistic-mixed-model","dir":"Articles","previous_headings":"Model complexity","what":"Logistic mixed model","title":"de Carvalho et al. 2021","text":"un-aggregated trial-level data use section E2_data, comes initial data preparation code chunk: apply Helmert/deviation-coded contrast matrix: specification object E2_data, add trial = \"Trial\" predict Target instead Prop. also add -subject random intercept formula: , CPA family = \"binomial\": now visualize results CPA_helm_complex CPA_helm side side: results largely , except largest, significant cluster identified TvsRI extends much complex CPA simple CPA. examine difference next.","code":"E2_data #> # A tibble: 69,123 × 5 #> Subject Trial Condition Time Target #> #> 1 200.asc 0 RightDislocated 0 0 #> 2 200.asc 0 RightDislocated 400 0 #> 3 200.asc 0 RightDislocated 450 1 #> 4 200.asc 0 RightDislocated 500 1 #> 5 200.asc 0 RightDislocated 550 1 #> 6 200.asc 0 RightDislocated 600 1 #> 7 200.asc 0 RightDislocated 650 1 #> 8 200.asc 0 RightDislocated 700 1 #> 9 200.asc 0 RightDislocated 750 1 #> 10 200.asc 0 RightDislocated 800 1 #> # ℹ 69,113 more rows E2_data$ConditionHelm <- E2_data$Condition contrasts(E2_data$ConditionHelm) <- condition_helm MASS::fractions(contrasts(E2_data$ConditionHelm)) #> RvsI TvsRI #> Intransitive -1/2 -1/3 #> RightDislocated 1/2 -1/3 #> Transitive 0 2/3 spec_helm_complex <- make_jlmer_spec( formula = Target ~ ConditionHelm + (1 | Subject), data = E2_data, subject = \"Subject\", trial = \"Trial\", time = \"Time\" ) spec_helm_complex #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: Target ~ 1 + ConditionHelmRvsI + ConditionHelmTvsRI + (1 | Subject) #> Predictors: #> ConditionHelm: ConditionHelmRvsI, ConditionHelmTvsRI #> Groupings: #> Subject: Subject #> Trial: Trial #> Time: Time #> Data: #> # A tibble: 69,123 × 6 #> Target ConditionHelmRvsI ConditionHelmTvsRI Subject Trial Time #> #> 1 0 0.5 -0.333 200.asc 0 0 #> 2 0 0.5 -0.333 200.asc 0 400 #> 3 1 0.5 -0.333 200.asc 0 450 #> # ℹ 69,120 more rows #> ──────────────────────────────────────────────────────────────────────────────── reset_rng_state() CPA_helm_complex <- clusterpermute( spec_helm_complex, family = \"binomial\", threshold = 1.5, nsim = 100, progress = FALSE ) CPA_helm_complex #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1.5) ──────────── ── #> ConditionHelmRvsI (n = 100) #> Mean (SD): 1.791 (31.12) #> Coverage intervals: 95% [-60.956, 80.601] #> ConditionHelmTvsRI (n = 100) #> Mean (SD): 2.704 (28.90) #> Coverage intervals: 95% [-54.636, 62.925] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1.5) ──────────────────────── ── #> ConditionHelmRvsI #> [2100, 3450]: 69.563 (p=0.0495) #> [4200, 5050]: 40.460 (p=0.1584) #> [5150, 5350]: 8.379 (p=0.7228) #> ConditionHelmTvsRI #> [350, 800]: 19.358 (p=0.4158) #> [900, 1050]: 7.519 (p=0.6535) #> [1250, 1350]: 4.980 (p=0.7228) #> [3100, 4450]: 56.770 (p=0.0792) #> [4650, 8000]: 181.165 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── clusters_annotation2 <- tidy(CPA_helm_complex$empirical_clusters) %>% mutate(contrast = gsub(\".*([TR])vs([RI]+)\", \"\\\\1 vs. \\\\2\", predictor)) %>% ggplot(aes(y = fct_rev(contrast))) + geom_segment( aes( x = start, xend = end, yend = contrast, color = pvalue < 0.05 ), linewidth = 8 ) + scale_color_manual(values = c(\"grey70\", \"steelblue\")) + scale_y_discrete() + scale_x_continuous(n.breaks = 9, limits = range(E2_data$Time)) + theme_minimal() + theme( legend.position = \"bottom\", axis.title = element_blank(), axis.text.y = element_text(face = \"bold\"), panel.border = element_rect(fill = NA), panel.grid.major.y = element_blank() ) clusters_annotation / clusters_annotation2 & guides(color = guide_none()) & plot_annotation(tag_levels = list(c(\"Simple\", \"Complex\")))"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/deCarvalho-et-al-2021.html","id":"comparison-of-cpas","dir":"Articles","previous_headings":"Model complexity","what":"Comparison of CPAs","title":"de Carvalho et al. 2021","text":"Looking timewise statistics computed simple vs. complex CPA tells us . figure plots information calls compute_timewise_statistics(): Whereas largest cluster starts emerge 5800ms CPA_helm, emerges much earlier 4650ms CPA_helm_complex. zoom region around 5800ms, see timewise statistics T vs. RI CPA_helm suddenly dip 1.5 threshold 5750ms: CPA better? existence dips spikes indicate problem, ’s consistent expectation simple CPA less robust variance. can inspect time-point model 5750ms two CPAs fitting : ’s standard way comparing goodness fit linear fixed-effects model logistic mixed-effects model fitted different data. complex model outperforms simple model classic metrics inspect glance(). doesn’t come surprise, differences largely driven number observations (nobs). Determining appropriate degree model complexity CPA beyond scope vignette, pursue discussion . Instead, conclude old wisdom: chain strong weakest link.","code":"# Compute the timewise statistics from the CPA specifications empirical_statistics <- bind_rows( tidy(compute_timewise_statistics(spec_helm)), tidy(compute_timewise_statistics(spec_helm_complex, family = \"binomial\")), .id = \"spec\" ) %>% mutate( CPA = c(\"Simple\", \"Complex\")[as.integer(spec)], Contrast = gsub(\".*([TR])vs([RI]+)\", \"\\\\1 vs. \\\\2\", predictor) ) # Time series plot of the statistics, with a line for each Helmert contrasts empirical_statistics_fig <- ggplot(empirical_statistics, aes(time, statistic)) + geom_line(aes(color = Contrast, linetype = CPA), linewidth = 1, alpha = .7) + geom_hline(yintercept = c(-1.5, 1.5), linetype = 2) + theme_classic() empirical_statistics_fig empirical_statistics_fig + geom_vline(xintercept = 5750) + coord_cartesian(xlim = 5750 + c(-500, 500), ylim = c(1, 2.5)) jlmer_simple_5750 <- to_jlmer( formula = Prop ~ ConditionHelm, data = E2_data_agg %>% filter(Time == 5750) ) jlmer_complex_5750 <- to_jlmer( formula = Target ~ ConditionHelm + (1 | Subject), family = \"binomial\", data = E2_data %>% filter(Time == 5750) ) glance(jlmer_simple_5750) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 72 4 0.223 7.51 -7.02 2.08 3.42 68 glance(jlmer_complex_5750) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 448 4 NA -301. 609. 626. 601. 444"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"background","dir":"Articles","previous_headings":"","what":"Background","title":"Garrison et al. 2020","text":"data comes eye-tracking study Garrison, Baudet, Breitfield, Aberman, & Bergelson (2020) “Familiarity plays small role noun comprehension 12-18 months”, way peekbankr corpus package. original study used growth curve model analyze children’s looks object familiar (“”) vs. unfamiliar (“”) different ages (median-split “younger” “older” groups). replicating specific analysis . Instead, follow Katie Von Holzen’s tutorial used dataset conduct simpler, cluster-based permutation analysis (CPA) differences two age groups proportion looks target. tutorial used eyetrackingR package data preparation CPA, vignette use dplyr jlmerclusterperm tasks, respectively. following set-code reads data prepares analysis: data ts_data_agg ’ll use replicate tutorial just 4 columns relevant cluster-permutation analysis: subject_id: Unique identifier subjects age: -participant factor variable (\"Younger\", \"Older\") time: continuous measure time 0-3975ms 25ms intervals prop: -subject proportion looks target object following replicates figure tutorial introduces cluster-permutation analysis:","code":"library(dplyr) remotes::install_github(\"langcog/peekbankr\", quiet = TRUE) library(peekbankr) # Load data aoi_timepoints <- get_aoi_timepoints(dataset_name = \"garrison_bergelson_2020\") administrations <- get_administrations(dataset_name = \"garrison_bergelson_2020\") # Pre-processing ## Filter for age groups of interest ps_data <- aoi_timepoints %>% left_join(administrations, by = \"administration_id\") %>% filter(!between(age, 14, 16)) %>% mutate(age_binned = ifelse(age < 14, \"Younger\", \"Older\")) ## Filter for time window of interest ts_window <- ps_data |> filter(t_norm >= 0, t_norm < 4000) ## Identify trials to exclude (trackloss in >25% of samples) to_exclude <- ts_window |> group_by(subject_id, trial_id) |> summarize(prop = mean(aoi == \"missing\"), .groups = \"drop\") |> filter(prop > 0.25) ## Exclude disqualifying trials and keep information relevant for CPA ts_data <- ts_window |> anti_join(to_exclude, by = c(\"subject_id\", \"trial_id\")) |> mutate( target = aoi == \"target\", missing = aoi == \"missing\", age = factor(age_binned, c(\"Younger\", \"Older\")) ) |> select(subject_id, age, trial_id, time = t_norm, target, missing) # Data of subject mean proportions of fixations to target ts_data_agg <- ts_data |> group_by(subject_id, age, time) |> summarize(prop = sum(target) / sum(!missing), .groups = \"drop\") # Data of trial-level fixations to target (0s and 1s) ts_data_trials <- ts_data %>% filter(!missing) %>% mutate(target = as.integer(target)) %>% select(subject_id, age, trial_id, time, target) ts_data_agg #> # A tibble: 2,560 × 4 #> subject_id age time prop #> #> 1 1413 Older 0 0.7 #> 2 1413 Older 25 0.733 #> 3 1413 Older 50 0.7 #> 4 1413 Older 75 0.7 #> 5 1413 Older 100 0.677 #> 6 1413 Older 125 0.677 #> 7 1413 Older 150 0.645 #> 8 1413 Older 175 0.645 #> 9 1413 Older 200 0.581 #> 10 1413 Older 225 0.548 #> # ℹ 2,550 more rows library(ggplot2) tutorial_fig <- ggplot(ts_data_agg, aes(x = time, y = prop, color = age)) + stat_smooth(method = \"loess\", span = 0.1, se = TRUE, aes(fill = age), alpha = 0.3) + theme_bw() + labs(y = \"Prop. of Target Looks\") + geom_hline(yintercept = .5, color = \"black\") + geom_vline(xintercept = 0, color = \"black\") + coord_cartesian(ylim = c(0, 1)) + theme( legend.position = \"bottom\", legend.title = element_blank() ) tutorial_fig"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"outline","dir":"Articles","previous_headings":"","what":"Outline","title":"Garrison et al. 2020","text":"case study vignette showcases four features CPA jlmerclusterperm: Prepping data CPA using make_jlmer_spec() CPA one go clusterpermute() modular, step--step approach CPA Using logistic regression trial-level data 1s 0s Load package start Julia instance jlmerclusterperm_setup() proceeding.","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"a-prepping-a-specification-object","dir":"Articles","previous_headings":"","what":"A) Prepping a specification object","title":"Garrison et al. 2020","text":"Conducting cluster-based permutation analysis jlmerclusterm starts creating special specification object, compiles information necessary CPA. function make_jlmer_spec() returns specification object class minimally requires two arguments: formula: R regression model formula data: data frame example, model prop using age predictor ts_data_agg data: printed display includes four pieces information specification object: Formula: model formula Julia syntax. , looks similar formula provided, contrasts “spelled ” (age becomes ageOlder) Predictors: list predictors form ( original : expanded ). case, one predictor age expanded ageOlder ’s treatment coded “Younger” reference level. Groupings: specified call make_jlmer_spec(). empty simple_spec provided formula data. Data: transformed dataframe whose columns terms expanded formula. Note default dummy coding applied discrete variable age - now ageOlder 0 “Younger” 1 “Older”. start bare specification object , lacks parts CPA, can use jlmerclusterperm’s simple interface Julia regression models. function jlmer() takes specification object input returns Julia regression model. model represents “global” model fitted entire data, collapsing across time. equivalent lm() model specifications: raise tangent jlmer() recommend sanity-checking design output global model prior using model specifications CPA. regression models fit Julia “hood”, want make sure output comparable expect R. global model can also tell upper bound model complexity. example, global model singular fit, time-wise models fitted subsets data likely . Functions CPA also take object require information time (calculate time-wise statistics) subject/trial (bootstrapped permutation). ts_data_agg data trial-level information summarized subject, just leave . trial argument can omitted object time sample/bin uniquely identified subject. Otherwise, trial column data uniquely identifies time values within subject (example, column items counterbalanced designed participant sees every item one variants).","code":"simple_spec <- make_jlmer_spec(formula = prop ~ 1 + age, data = ts_data_agg) simple_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: prop ~ 1 + ageOlder #> Predictors: #> age: ageOlder #> Data: #> # A tibble: 2,560 × 2 #> prop ageOlder #> #> 1 0.7 1 #> 2 0.733 1 #> 3 0.7 1 #> # ℹ 2,557 more rows #> ──────────────────────────────────────────────────────────────────────────────── global_jmod <- jlmer(simple_spec) tidy(global_jmod) #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.523 0.00387 135. 0 #> 2 ageOlder 0.0984 0.00548 18.0 3.45e-72 library(broom) # for the `tidy()` method for `lm()` global_rmod <- lm(formula = prop ~ 1 + age, data = ts_data_agg) tidy(global_rmod) #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.523 0.00387 135. 0 #> 2 ageOlder 0.0984 0.00548 18.0 4.45e-68 ts_data_agg_spec <- make_jlmer_spec( formula = prop ~ 1 + age, data = ts_data_agg, subject = \"subject_id\", time = \"time\", trial = NULL # This is the default value ) ts_data_agg_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: prop ~ 1 + ageOlder #> Predictors: #> age: ageOlder #> Groupings: #> Subject: subject_id #> Trial: #> Time: time #> Data: #> # A tibble: 2,560 × 4 #> prop ageOlder subject_id time #> #> 1 0.7 1 1413 0 #> 2 0.733 1 1413 25 #> 3 0.7 1 1413 50 #> # ℹ 2,557 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"b-cpa-with-clusterpermute","dir":"Articles","previous_headings":"","what":"B) CPA with clusterpermute()","title":"Garrison et al. 2020","text":"clusterpermute() function runs CPA one go, using information object. addition specification object, must also supply (t-value) threshold number simulations run: result list two elements: object holds information distribution bootstrapped cluster-mass statistics random permutations data: object holds per-predictor information clusters detected, p-values significance testing null: output explained next section, presented purposes simply note results similar reported Katie Von Holzen’s tutorial, copied . minor numerical differences due stochastic nature permutation test ’re comparing t-tests R regression models Julia.","code":"CPA_agg <- clusterpermute( ts_data_agg_spec, threshold = 2, nsim = 100, progress = FALSE # Suppress printing of progress for vignette ) CPA_agg #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> ageOlder (n = 100) #> Mean (SD): -0.844 (29.71) #> Coverage intervals: 95% [-46.664, 71.698] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.0396) #> [2675, 2775]: 11.017 (p=0.4356) #> [3500, 3900]: 46.295 (p=0.0990) #> ──────────────────────────────────────────────────────────────────────────────── #> ## Summary of Clusters ====== #> ## Cluster Direction SumStatistic StartTime EndTime Probability #> ## 1 1 Positive 74.723687 850 1550 0.051 #> ## 2 2 Positive 6.905622 2725 2800 0.405 #> ## 3 3 Positive 46.294644 3500 3925 0.111"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"c-a-step-by-step-approach","dir":"Articles","previous_headings":"","what":"C) A step-by-step approach","title":"Garrison et al. 2020","text":"clusterpermute() function showcased consists five parts, also standalone functions package: compute_timewise_statistics() extract_empirical_clusters() permute_timewise_statistics() extract_null_cluster_dists() calculate_clusters_pvalues() functions correspond algorithmic steps statistical test. clusterpermute() obviates need think individual components, knowing gives greater flexibility procedure. section walks CPA step--step using functions.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"empirical-clusters","dir":"Articles","previous_headings":"C) A step-by-step approach","what":"1) Empirical clusters","title":"Garrison et al. 2020","text":"compute_timewise_statistics() function takes object returns timewise statistics form matrix row predictor column time point: example see 25ms, t-value age -0.127. consistent figure tutorial replicated show two lines mostly overlapping. Just demonstration purposes, use to_jlmer() quickly fit model using formula just data 25ms. see t-value age model exactly saw : construct empirical clusters timewise statistics. Clusters defined continuous span statistics sign whose absolute value passes certain threshold. extract_empirical_clusters() function takes timewise statistics threshold just : t-value threshold 2, detect three clusters data main effect age (ageOlder). numbers brackets show span cluster number right show sum t-values cluster (.k.. cluster-mass statistic). can collect object data frame tidy(): add figure : point know whether size clusters observe (un)likely emerge chance. calculate probability, need simulate null distribution.","code":"empirical_statistics <- compute_timewise_statistics(ts_data_agg_spec) dim(empirical_statistics) #> [1] 1 160 empirical_statistics[1, 1:5, drop = FALSE] # First 5 time points #> Time #> Predictor 0 25 50 75 100 #> ageOlder 0.1166025 -0.1273179 -0.2387778 -0.4022095 -0.2804397 to_jlmer(prop ~ 1 + age, data = filter(ts_data_agg, time == 25)) %>% tidy() %>% filter(term == \"ageOlder\") %>% pull(statistic) #> [1] -0.1273179 empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 #> [2675, 2775]: 11.017 #> [3500, 3900]: 46.295 #> ──────────────────────────────────────────────────────────────────────────────── empirical_clusters_df <- tidy(empirical_clusters) empirical_clusters_df #> # A tibble: 3 × 6 #> predictor id start end length sum_statistic #> #> 1 ageOlder 1 850 1525 28 74.7 #> 2 ageOlder 2 2675 2775 5 11.0 #> 3 ageOlder 3 3500 3900 17 46.3 tutorial_fig + geom_segment( aes(x = start, xend = end, y = -Inf, yend = -Inf), linewidth = 10, color = \"steelblue\", inherit.aes = FALSE, data = empirical_clusters_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"Cluster\", id) ), vjust = -2, inherit.aes = FALSE, data = empirical_clusters_df )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"null-distribution","dir":"Articles","previous_headings":"C) A step-by-step approach","what":"2) Null distribution","title":"Garrison et al. 2020","text":"permute_timewise_statistics() function takes specification object number bootstrapped permutation data simulate (via nsim argument). returned 3-dimensional array like output compute_timewise_statistics() except additional dimension representing simulation: permutation algorithm preserves grouping structures data specified specification object. example, testing -subject predictor like age, age groups participants random swapped preserving temporal structure data. null distribution collection largest cluster-mass statistic simulated data. clusters detected simulation, contributes cluster-mass zero null. use extract_null_cluster_dists() construct null distribution, using threshold value 2 allow comparison empirical clusters: printed, returned object shows summary statistics. can use tidy() collect samples null data frame: null distribution can plotted like : Now left test probability observing cluster-mass statistics extreme detected clusters empirical_clusters, using null_cluster_dists reference.","code":"null_statistics <- permute_timewise_statistics(ts_data_agg_spec, nsim = 100) # simulation by time by predictor dim(null_statistics) #> [1] 100 160 1 # First 5 time points of the first three simulations null_statistics[1:3, 1:5, 1, drop = FALSE] #> , , Predictor = ageOlder #> #> Time #> Sim 0 25 50 75 100 #> 001 -0.1037696 0.1387838 0.4583545 0.4337538 0.69136613 #> 002 1.4358330 1.2745381 1.2985888 0.9111359 1.16401311 #> 003 0.2545470 0.3885999 0.3374939 0.3200497 0.05603617 null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> ageOlder (n = 100) #> Mean (SD): -6.150 (40.29) #> Coverage intervals: 95% [-114.564, 74.425] #> ──────────────────────────────────────────────────────────────────────────────── null_cluster_dists_df <- tidy(null_cluster_dists) null_cluster_dists_df #> # A tibble: 100 × 6 #> predictor start end length sum_statistic sim #> #> 1 ageOlder NA NA NA 0 001 #> 2 ageOlder NA NA NA 0 002 #> 3 ageOlder 2175 2225 3 -6.37 003 #> 4 ageOlder NA NA NA 0 004 #> 5 ageOlder 3425 3475 3 -6.67 005 #> 6 ageOlder NA NA NA 0 006 #> 7 ageOlder NA NA NA 0 007 #> 8 ageOlder NA NA NA 0 008 #> 9 ageOlder NA NA NA 0 009 #> 10 ageOlder 1525 2475 39 -114. 010 #> # ℹ 90 more rows plot( density(null_cluster_dists_df$sum_statistic), main = \"Null distribution of cluster-mass statistics\" )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"significance-test","dir":"Articles","previous_headings":"C) A step-by-step approach","what":"3) Significance test","title":"Garrison et al. 2020","text":"calculate_clusters_pvalues() function computes p-value empirical cluster returns augmented object: , p-values slightly different test stochastic. separate vignette covers issues replicability, seed, rng. collected tidy(), p-values added column: Note reporting pvalues, ’s customary add 1 numerator denominator. observed data, however unlikely, one particular arrangement (permutation) data. controlled via add1 argument - false default calculate_clusters_pvalues() true clusterpermute(). now come full circle: pvalue-augmented object, along object, also clusterpermute() function returned .","code":"empirical_clusters_tested <- calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) empirical_clusters_tested #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.1100) #> [2675, 2775]: 11.017 (p=0.4000) #> [3500, 3900]: 46.295 (p=0.1700) #> ──────────────────────────────────────────────────────────────────────────────── tidy(empirical_clusters_tested) #> # A tibble: 3 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 ageOlder 1 850 1525 28 74.7 0.11 #> 2 ageOlder 2 2675 2775 5 11.0 0.4 #> 3 ageOlder 3 3500 3900 17 46.3 0.17 calculate_clusters_pvalues(empirical_clusters, null_cluster_dists, add1 = TRUE) #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.1188) #> [2675, 2775]: 11.017 (p=0.4059) #> [3500, 3900]: 46.295 (p=0.1782) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"aggregated-vs--trial-level-data","dir":"Articles","previous_headings":"D) As logistic regression","what":"Aggregated vs. trial-level data","title":"Garrison et al. 2020","text":"data using far aggregated data collapsed across trials within subject. words, exactly one measure proportion looks target per time point within subject: possible also fit logistic regression model trial-level measure “hits” (1s 0s) target area interest. un-aggregated, trial-level data comes data preparation code top vignette. demonstrate relationship two data sets: specification object fitting logistic regression models trial-level data requires two changes original: formula must predict binary target variable, instead prop trial grouping must specified uniquely identify time point within subject following appropriate object CPA using logistic regression: , sanity check global model using family = \"binomial\": estimates similar global_jmod previous section, log-odds.","code":"nrow(ts_data_agg) #> [1] 2560 ts_data_agg %>% distinct(subject_id, time) %>% nrow() #> [1] 2560 ts_data_trials #> # A tibble: 62,466 × 5 #> subject_id age trial_id time target #> #> 1 1413 Older 2659 0 1 #> 2 1413 Older 2659 25 1 #> 3 1413 Older 2659 50 1 #> 4 1413 Older 2659 75 1 #> 5 1413 Older 2659 100 1 #> 6 1413 Older 2659 125 1 #> 7 1413 Older 2659 150 1 #> 8 1413 Older 2659 175 1 #> 9 1413 Older 2659 200 1 #> 10 1413 Older 2659 225 1 #> # ℹ 62,456 more rows identical( ts_data_agg, ts_data_trials %>% group_by(subject_id, age, time) %>% summarize(prop = mean(target), .groups = \"drop\") ) #> [1] TRUE ts_data_trial_spec <- make_jlmer_spec( formula = target ~ 1 + age, data = ts_data_trials, subject = \"subject_id\", trial = \"trial_id\", time = \"time\" ) ts_data_trial_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: target ~ 1 + ageOlder #> Predictors: #> age: ageOlder #> Groupings: #> Subject: subject_id #> Trial: trial_id #> Time: time #> Data: #> # A tibble: 62,466 × 5 #> target ageOlder subject_id trial_id time #> #> 1 1 1 1413 2659 0 #> 2 1 1 1413 2659 25 #> 3 1 1 1413 2659 50 #> # ℹ 62,463 more rows #> ──────────────────────────────────────────────────────────────────────────────── global_jmod_binom <- jlmer(ts_data_trial_spec, family = \"binomial\") tidy(global_jmod_binom) #> # A tibble: 2 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 0.113 0.0121 9.31 1.24e- 20 #> 2 ageOlder 0.392 0.0164 23.9 1.08e-126"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Garrison-et-al-2020.html","id":"repeating-the-cpa-with-logistic-regression","dir":"Articles","previous_headings":"D) As logistic regression","what":"Repeating the CPA with logistic regression","title":"Garrison et al. 2020","text":"repeat procedure use new specification object set family = \"binomial\": Instead 3 clusters, now detect 5. can annotate detected clusters time series data : differences see aggregated vs. trial-level data completely expected follow straightforwardly following two facts: cluster-mass statistic derived size t-values time point, effect size. virtue unaggregated ts_data_trials data number observations, t-values overall inflated difference means (also see output global_jmod vs. global_jmod_binom well). Whereas aggregated data reflects complete pooling trial-level variation, binomial model trial-level data performs pooling treats trial independent. Thus, certain item experiment show outlier effect one time point, example, strongly influence mass cluster spanning given time point. first issue (inflated statistics) trivial just takes matter adjusting threshold. fact, good “stretching ” t-statistics, get higher resolution shape timewise statistics. However. second issue (pooling) troubling can warp fundamental shape timewise statistics. diagnose factors visualizing timewise statistics. also plot timewise statistics aggregated data comparison: see trial-level binomial model overall inflated t-values, well pronounced “peaks” (specifically maximum 1300ms). middle-ground, partial pooling approaching using mixed-effects models CPA showcased case study vignettes. present purposes, raise threshold value (2.8) gets us close detecting least largest cluster empirical_statistics, spanning 850ms-1525ms: run cluster permutation test threshold just largest cluster: gives us pvalue cluster spanning 850ms 1525ms similar one reported aggregated data: appear converge higher nsim data set. also time 1000-simulation runs showcase speed jlmerclusterperm. Note use set_rng_state(), makes two CPAs generate permutations observed data.","code":"empirical_statistics_binom <- compute_timewise_statistics(ts_data_trial_spec, family = \"binomial\") empirical_clusters_binom <- extract_empirical_clusters(empirical_statistics_binom, threshold = 2) empirical_clusters_binom #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [650, 725]: 8.727 #> [800, 1575]: 112.199 #> [1775, 1875]: 12.764 #> [2600, 2825]: 23.733 #> [3475, 3975]: 62.714 #> ──────────────────────────────────────────────────────────────────────────────── empirical_clusters_binom_df <- tidy(empirical_clusters_binom) tutorial_fig + geom_segment( aes(x = start, xend = end, y = -Inf, yend = -Inf), linewidth = 10, color = \"steelblue\", inherit.aes = FALSE, data = empirical_clusters_binom_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"Cluster\", id) ), vjust = -2, inherit.aes = FALSE, data = empirical_clusters_binom_df ) timewise_statistics_fig <- ggplot(mapping = aes(time, statistic)) + geom_hline(aes(yintercept = 0)) + geom_line( aes(color = \"aggregated\"), linewidth = 1.5, data = tidy(empirical_statistics) ) + geom_line( aes(color = \"trial-level\"), linewidth = 1.5, data = tidy(empirical_statistics_binom) ) timewise_statistics_fig + geom_vline( aes(xintercept = time), data = tidy(empirical_statistics_binom) %>% slice(which.max(abs(statistic))) ) empirical_clusters_binom2 <- extract_empirical_clusters(empirical_statistics_binom, threshold = 2.8) empirical_clusters_binom2 #> ── Empirical clusters (t > 2.8) ──────────────────────── ── #> ageOlder #> [850, 1525]: 102.450 #> [3500, 3875]: 51.365 #> ──────────────────────────────────────────────────────────────────────────────── CPA_trial <- clusterpermute( ts_data_trial_spec, family = \"binomial\", threshold = 2.8, nsim = 100, top_n = 1, # Just test the largest cluster progress = FALSE # Suppress printing of progress for vignette ) CPA_trial$empirical_clusters #> ── Empirical clusters (t > 2.8) ──────────────────────── ── #> ageOlder #> [850, 1525]: 102.450 (p=0.0297) #> ──────────────────────────────────────────────────────────────────────────────── CPA_agg$empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.0396) #> [2675, 2775]: 11.017 (p=0.4356) #> [3500, 3900]: 46.295 (p=0.0990) #> ──────────────────────────────────────────────────────────────────────────────── system.time({ set_rng_state(123L) CPA_agg_1000 <- clusterpermute( ts_data_agg_spec, threshold = 2, nsim = 1000, top_n = 1, progress = FALSE ) }) #> user system elapsed #> 0.27 0.58 17.53 CPA_agg_1000 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> ageOlder (n = 1000) #> Mean (SD): 0.078 (40.29) #> Coverage intervals: 95% [-76.748, 82.717] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> ageOlder #> [850, 1525]: 74.724 (p=0.0629) #> ──────────────────────────────────────────────────────────────────────────────── system.time({ set_rng_state(123L) CPA_trial_1000 <- clusterpermute( ts_data_trial_spec, family = \"binomial\", threshold = 2.8, nsim = 1000, top_n = 1, progress = FALSE ) }) #> user system elapsed #> 0.14 0.20 31.39 CPA_trial_1000 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2.8) ──────────── ── #> ageOlder (n = 1000) #> Mean (SD): -0.045 (55.09) #> Coverage intervals: 95% [-119.239, 125.506] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2.8) ──────────────────────── ── #> ageOlder #> [850, 1525]: 102.450 (p=0.0689) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"background","dir":"Articles","previous_headings":"","what":"Background","title":"Geller et al. 2020","text":"data comes lexical decision task using pupilometry (specifically, measure pupil dilation) study differences processing difficulty print vs. cursive script. data available part gazer package Geller, Winn, Mahr, & Mirman 2020. follow Jason Geller’s tutorial used clusterperm package conduct cluster-based permutation analysis (CPA) differences pupil size print cursive conditions. data package (cursive_agg_data) looks slightly different used tutorial, ensure full reproducibility use exact data tutorial used: data comes prepared analysis box. following columns relevant analysis: subject: Unique identifier subjects script: within-participant factor variable (\"cursive\", \"print\") time: continuous measure time 0-2500ms 100ms intervals aggbaseline: response variable representing normalized (“baseline-corrected”) pupil size name cusrive_agg_data variable suggests, data aggregated within subject, collapsing across trials. following reproduces figure tutorial: within-participant predictors like script, permutation algorithm randomly swap labels condition trials within subject. preserves temporal structure trial-level data (swapping trial-level grouping) well subject-level grouping structure (swapping trials across participants).","code":"library(dplyr) cursive_agg_data <- as_tibble(read.csv(\"https://raw.githubusercontent.com/jgeller112/drjasongeller/main/content/blog/2020-07-10-CBPT/blog_data.csv\")) cursive_agg_data #> # A tibble: 2,184 × 5 #> X subject script timebins aggbaseline #> #> 1 1 10b.edf cursive 0 36.9 #> 2 2 10b.edf cursive 100 27.5 #> 3 3 10b.edf cursive 200 19.6 #> 4 4 10b.edf cursive 300 12.6 #> 5 5 10b.edf cursive 400 5.60 #> 6 6 10b.edf cursive 500 0.968 #> 7 7 10b.edf cursive 600 -5.98 #> 8 8 10b.edf cursive 700 -5.96 #> 9 9 10b.edf cursive 800 3.01 #> 10 10 10b.edf cursive 900 12.9 #> # ℹ 2,174 more rows library(ggplot2) #> Warning: package 'ggplot2' was built under R version 4.3.1 script_fig <- ggplot(cursive_agg_data, aes(timebins, aggbaseline)) + stat_summary(aes(linetype = script), geom = \"line\", linewidth = 1) script_fig"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"outline","dir":"Articles","previous_headings":"","what":"Outline","title":"Geller et al. 2020","text":"case study vignette showcases five features CPA jlmerclusterperm: Prepping data CPA using make_jlmer_spec() clusterpermute() default statistic = \"t\" comparison “t” “chisq” Specifying random effects Contrast coding Load package start Julia instance jlmerclusterperm_setup() proceeding.","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"a-prepping-a-specification-object","dir":"Articles","previous_headings":"","what":"A) Prepping a specification object","title":"Geller et al. 2020","text":"start simple specification object model aggbaseline using script predictor script_fig data: sanity check, fit global model data… …check ’s comparable expect lm(): full specification object CPA must also declare grouping structures present data. rule thumb every observation (row) data must uniquely identified combination columns subject, trial, time. CPA time series data time must always specified, subject must also always specified permutation algorithm respect subject-level grouping data. cursive_agg_data data collapses across trials within subject, column trial. However, leave trial argument unspecified observations uniquely identified subject time alone. instead 2 rows per subject-time combination, one script condition: Therefore need “dummy” column trial distinctly mark “cursive” vs. “script” trials. save new data cursive_agg: makes observations uniquely identified columns subject, trial, time: final specification object looks like following:","code":"simple_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script, data = cursive_agg_data ) simple_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: aggbaseline ~ 1 + scriptprint #> Predictors: #> script: scriptprint #> Data: #> # A tibble: 2,184 × 2 #> aggbaseline scriptprint #> #> 1 36.9 0 #> 2 27.5 0 #> 3 19.6 0 #> # ℹ 2,181 more rows #> ──────────────────────────────────────────────────────────────────────────────── jlmer(simple_spec) #> #> ──────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ──────────────────────────────────────────────────────────────────────── #> (Intercept) 46.7221 2.79622 16.71 <1e-61 41.2417 52.2026 #> scriptprint -10.2582 3.95445 -2.59 0.0095 -18.0088 -2.5076 #> ──────────────────────────────────────────────────────────────────────── summary(lm(formula = aggbaseline ~ 1 + script, data = cursive_agg_data))$coefficients #> Estimate Std. Error t value Pr(>|t|) #> (Intercept) 46.72214 2.796221 16.709029 4.512338e-59 #> scriptprint -10.25818 3.954454 -2.594083 9.547816e-03 cursive_agg_data %>% count(subject, timebins) #> # A tibble: 1,092 × 3 #> subject timebins n #> #> 1 10b.edf 0 2 #> 2 10b.edf 100 2 #> 3 10b.edf 200 2 #> 4 10b.edf 300 2 #> 5 10b.edf 400 2 #> 6 10b.edf 500 2 #> 7 10b.edf 600 2 #> 8 10b.edf 700 2 #> 9 10b.edf 800 2 #> 10 10b.edf 900 2 #> # ℹ 1,082 more rows cursive_agg <- cursive_agg_data %>% mutate(trial_type = paste0(script, \"_agg\")) cursive_agg #> # A tibble: 2,184 × 6 #> X subject script timebins aggbaseline trial_type #> #> 1 1 10b.edf cursive 0 36.9 cursive_agg #> 2 2 10b.edf cursive 100 27.5 cursive_agg #> 3 3 10b.edf cursive 200 19.6 cursive_agg #> 4 4 10b.edf cursive 300 12.6 cursive_agg #> 5 5 10b.edf cursive 400 5.60 cursive_agg #> 6 6 10b.edf cursive 500 0.968 cursive_agg #> 7 7 10b.edf cursive 600 -5.98 cursive_agg #> 8 8 10b.edf cursive 700 -5.96 cursive_agg #> 9 9 10b.edf cursive 800 3.01 cursive_agg #> 10 10 10b.edf cursive 900 12.9 cursive_agg #> # ℹ 2,174 more rows cursive_agg %>% count(subject, timebins, trial_type) %>% distinct(n) #> # A tibble: 1 × 1 #> n #> #> 1 1 cursive_agg_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script, data = cursive_agg, subject = \"subject\", trial = \"trial_type\", time = \"timebins\" ) cursive_agg_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: aggbaseline ~ 1 + scriptprint #> Predictors: #> script: scriptprint #> Groupings: #> Subject: subject #> Trial: trial_type #> Time: timebins #> Data: #> # A tibble: 2,184 × 5 #> aggbaseline scriptprint subject trial_type timebins #> #> 1 36.9 0 10b.edf cursive_agg 0 #> 2 27.5 0 10b.edf cursive_agg 100 #> 3 19.6 0 10b.edf cursive_agg 200 #> # ℹ 2,181 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"b-cpa-with-default-statistic-t","dir":"Articles","previous_headings":"","what":"B) CPA with default statistic = \"t\"","title":"Geller et al. 2020","text":"CPA output original tutorial (using 100 simulations) copied comparison: Using threshold 2 default statistic = \"t\", clusterpermute() returns following: results look different original. due following: cluster-mass statistic (cms) lower. ran clusterpermute() default statistic = \"t\". Different clusters identified. simple model account subject-level variation, whereas original error term formula aggbaseline ~ script + Error(subject). sign cluster reversed. defaults aov() used original tutorial different default contrast coding regression model. now address issue turn, building CPA closely replicate results tutorial.","code":"#> #> effect b0 b1 sign cms p #> #> script 0 100 1 10.121215 0.1584158 #> #> script 900 1000 -1 8.756152 0.1782178 #> #> script 1900 2500 1 82.198279 0.0099010 set_rng_state(123L) clusterpermute( cursive_agg_spec, statistic = \"t\", # Default value spelled out threshold = 2, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> scriptprint (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> scriptprint #> [2000, 2300]: -8.748 (p=0.0099) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"c-comparing-t-vs--chisq","dir":"Articles","previous_headings":"","what":"C) Comparing “t” vs. “chisq”","title":"Geller et al. 2020","text":"original tutorial used clusterperm::cluster_nhds() conduct CPA, fits ANOVA models time. , timewise statistic used chi-squared threshold determined p-value chi-squared statistic. Using chi-squared statistics p-value threshold also supported clusterpermute() using statistic = \"chisq\" (instead default \"t\"): returns cluster now numerically larger cluster-mass statistic, expected (chi-squared asymptotic t^2). , compare shape timewise statistics “t” threshold 2 “chisq” threshold p=0.05: find clusters identified 2000ms-2300ms “t” “chisq”, peaks pronounced “chisq”. differences two inconsequential example, may produce different results cases. chi-squared statistic (jlmerclusterperm computes via likelihood ratio test) often preferred testing single parameters makes less assumptions tend robust (glmmFAQ). goodness--fit tests level predictor model formula, “chisq” less interpretable multi-level predictors (k-1 > 1). teasing apart contribution individual levels multi-level predictor, using statistic = \"t\" appropriate.","code":"set_rng_state(123L) clusterpermute( cursive_agg_spec, statistic = \"chisq\", threshold = 0.05, # Threshold is now the p-value of the chi-squared statistic nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [2000, 2300]: -19.085 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── timewise_ts <- compute_timewise_statistics(cursive_agg_spec, statistic = \"t\") timewise_chisqs <- compute_timewise_statistics(cursive_agg_spec, statistic = \"chisq\") library(ggplot2) timewise_fig <- ggplot(mapping = aes(x = time, y = statistic)) + geom_line( aes(color = \"fixed-t\"), linewidth = 1.5, data = tidy(timewise_ts) ) + geom_line( aes(color = \"fixed-chisq\"), linewidth = 1.5, data = tidy(timewise_chisqs) ) + geom_hline( aes(yintercept = c(-2, 2, qchisq(.95, df = 1), -qchisq(.95, df = 1))), color = rep(c(\"#00BFC4\", \"#F8766D\"), each = 2), linetype = 2 ) + scale_color_manual( values = setNames( c(\"#E69F00\", \"#56B4E9\", \"#009E73\", \"#F0E442\"), c(\"fixed-t\", \"fixed-chisq\", \"re-intercept-chisq\", \"re-max-chisq\") ) ) timewise_fig"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"d-specifying-random-effects","dir":"Articles","previous_headings":"","what":"D) Specifying random effects","title":"Geller et al. 2020","text":"biggest missing component point subject random effects, original tutorial captures via error term Error(subject). isn’t strict equivalent regression, specifying random intercepts subject (1 | subject) gets us close: repeat results original tutorial comparison: now plot timewise statistics random intercept models: point may wonder whether results change much used maximal model random effects structure (1 + script | subject). first create another specification object maximal formula: compute timewise statistics: find chisq statistics maximal model virtually identical parsimonious, intercept-model (two lines overlap): Parsimony incredibly important simulation - jlmerclusterperm fast, model complexity still major bottleneck. adding extra random effect terms (random slope correlation) negligible effects statistic, removing likely inconsequential CPA . show following CPA run, uses maximal model. Notice results identical intercept-re_CPA substantially slower: run maximal mixed model CPA purely demonstration purposes. random effects structure actually unidentifiable given aggregated data, point time 2 observations subject (mean condition). Lastly, back figure data, annotate clusters detected chisq p-value threshold 0.05 random intercept CPA:","code":"cursive_agg_spec_re <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 | subject), data = cursive_agg, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) set_rng_state(123L) system.time({ re_CPA <- clusterpermute( cursive_agg_spec_re, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) }) #> user system elapsed #> 0.06 0.04 12.13 re_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): 1.132 (12.58) #> Coverage intervals: 95% [-21.081, 32.649] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: -9.770 (p=0.1881) #> [900, 1000]: 8.521 (p=0.2475) #> [1900, 2500]: -73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── #> #> effect b0 b1 sign cms p #> #> script 0 100 1 10.121215 0.1584158 #> #> script 900 1000 -1 8.756152 0.1782178 #> #> script 1900 2500 1 82.198279 0.0099010 timewise_chisqs_re <- compute_timewise_statistics(cursive_agg_spec_re, statistic = \"chisq\") timewise_fig_re <- timewise_fig + geom_line( aes(color = \"re-intercept-chisq\"), linewidth = 1.5, data = tidy(timewise_chisqs_re) ) timewise_fig_re cursive_agg_spec_re_max <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 + script | subject), data = cursive_agg, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) timewise_chisqs_re_max <- compute_timewise_statistics(cursive_agg_spec_re_max, statistic = \"chisq\") #> ℹ 1 singular fit (3.85%). timewise_fig_re + geom_line( aes(color = \"re-max-chisq\"), linewidth = 1.5, data = tidy(timewise_chisqs_re_max) ) set_rng_state(123L) system.time({ re_max_CPA <- clusterpermute( cursive_agg_spec_re_max, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) }) #> user system elapsed #> 0.05 0.03 12.99 re_max_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): 1.132 (12.58) #> Coverage intervals: 95% [-21.081, 32.649] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: -9.770 (p=0.1881) #> [900, 1000]: 8.521 (p=0.2475) #> [1900, 2500]: -73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── empirical_clusters_df <- tidy(extract_empirical_clusters(timewise_chisqs_re, threshold = 0.05)) script_fig + geom_segment( aes( x = start, xend = end, y = -Inf, yend = -Inf, color = factor(sign(sum_statistic)) ), linewidth = 10, inherit.aes = FALSE, data = empirical_clusters_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"Cluster\", id) ), vjust = -2, inherit.aes = FALSE, data = empirical_clusters_df ) + labs(color = \"Sign of cluster\")"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/Geller-et-al-2020.html","id":"e-contrast-coding","dir":"Articles","previous_headings":"","what":"E) Contrast coding","title":"Geller et al. 2020","text":"last piece puzzle flipped sign effect. Whereas detect negative cluster line cursive line print, original tutorial reports positive cluster (output repeated ): Fixing trivial - just takes different choice contrast. example, can flip levels factor make “print” reference level: also sum-code set “print” -1 “cursive” 1: CPA operates domain timewise statistics (effect size), whether difference “print” “cursive” expressed unit 1 (treatment coding) 2 (contrast coding) bearing CPA. conclude, re-run time 1000-simulation CPA using intercept-model new treatment coding script “print” reference level. Finally, annotating just significant cluster figure:","code":"#> #> effect b0 b1 sign cms p #> #> script 0 100 1 10.121215 0.1584158 #> #> script 900 1000 -1 8.756152 0.1782178 #> #> script 1900 2500 1 82.198279 0.0099010 rev_contrast_df <- cursive_agg rev_contrast_df$script <- factor(rev_contrast_df$script, levels = c(\"print\", \"cursive\")) contrasts(rev_contrast_df$script) #> cursive #> print 0 #> cursive 1 reverse_contrast_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 | subject), data = rev_contrast_df, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) set_rng_state(123L) clusterpermute( reverse_contrast_spec, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): -1.132 (12.58) #> Coverage intervals: 95% [-32.649, 21.081] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: 9.770 (p=0.1881) #> [900, 1000]: -8.521 (p=0.2475) #> [1900, 2500]: 73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── sum_contrast_df <- cursive_agg sum_contrast_df$script <- factor(sum_contrast_df$script) contrasts(sum_contrast_df$script) <- contr.sum(2) contrasts(sum_contrast_df$script) #> [,1] #> cursive 1 #> print -1 sum_contrast_spec <- make_jlmer_spec( formula = aggbaseline ~ 1 + script + (1 | subject), data = sum_contrast_df, subject = \"subject\", time = \"timebins\", trial = \"trial_type\" ) set_rng_state(123L) clusterpermute( sum_contrast_spec, statistic = \"chisq\", threshold = 0.05, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 100, df = 1) #> Mean (SD): -1.132 (12.58) #> Coverage intervals: 95% [-32.649, 21.081] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: 9.770 (p=0.1881) #> [900, 1000]: -8.521 (p=0.2475) #> [1900, 2500]: 73.440 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── set_rng_state(123L) system.time({ final_CPA <- clusterpermute( reverse_contrast_spec, statistic = \"chisq\", threshold = 0.05, nsim = 1000, progress = FALSE ) }) #> user system elapsed #> 0.03 0.11 5.28 final_CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (chisq p < 0.05) ───── ── #> script (n = 1000, df = 1) #> Mean (SD): -0.107 (16.50) #> Coverage intervals: 95% [-36.966, 41.530] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (chisq p < 0.05) ───────────────── ── #> script (df = 1) #> [0, 100]: 9.770 (p=0.2108) #> [900, 1000]: -8.521 (p=0.2358) #> [1900, 2500]: 73.440 (p=0.0090) #> ──────────────────────────────────────────────────────────────────────────────── signif_clusters_df <- tidy(final_CPA$empirical_clusters) %>% filter(pvalue < 0.05) signif_clusters_df #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 script 3 1900 2500 7 73.4 0.00899 script_fig + geom_segment( aes(x = start, xend = end, y = -Inf, yend = -Inf), color = \"steelblue\", linewidth = 10, inherit.aes = FALSE, data = signif_clusters_df ) + geom_text( aes( y = -Inf, x = start + (end - start) / 2, label = paste(\"p =\", signif(pvalue, 3)) ), vjust = -2, inherit.aes = FALSE, data = signif_clusters_df )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"overview-of-cpa","dir":"Articles","previous_headings":"","what":"Overview of CPA","title":"Introduction to jlmerclusterperm","text":"Cluster-based permutation analysis (CPA) simulation-based, non-parametric statistical test difference groups time series. suitable analyzing densely-sampled time data (EEG eye-tracking research) research hypothesis often specified existence effect (e.g., predicted higher-order cognitive mechanisms) agnostic temporal details effect (precise moment emergence). CPA formalizes two intuitions means difference groups: countable unit difference (.e., cluster) contiguous, uninterrupted span sufficiently large differences time point (determined via threshold). degree extremity cluster (.e., cluster-mass statistic) measure sensitive magnitude difference, variability, sample size (e.g., t-statistic regression). CPA procedure identifies empirical clusters time series data tests significance observed cluster-mass statistics bootstrapped permutations data. bootstrapping procedure samples “null” (via random shuffling condition labels), yielding distribution cluster-mass statistics emerging chance. statistical significance cluster probability observing cluster-mass statistic extreme cluster’s simulated null distribution.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"package-design","dir":"Articles","previous_headings":"","what":"Package design","title":"Introduction to jlmerclusterperm","text":"jlmerclusterperm provides wholesale piecemeal approach conducting CPA. main workhorse function package clusterpermute(), composed five smaller functions called internally succession. smaller functions representing algorithmic steps CPA also exported, allow control procedure (e.g., debugging diagnosing CPA run). See function documentation .","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"organization-of-vignettes","dir":"Articles","previous_headings":"","what":"Organization of vignettes","title":"Introduction to jlmerclusterperm","text":"package vignettes roughly divided two groups: topics case studies. researcher data ready analysis, recommended go case studies first pluck relevant bits desired analysis order increasing complexity, case studies : Garrison et al. 2020, introduces package’s functions running CPA, demonstrating wholesale piecemeal approaches. also compares use linear vs. logistic regression calculation timewise statistics. Geller et al. 2020, demonstrates use random effects regression contrasts CPA specification. also compares use t vs. chisq timewise statistics. de Carvalho et al. 2021, showcases using custom regression contrasts test relationship multiple levels factor. also discusses issues surrounding complexity timewise regression model. topics cover general package features: Tidying output: tidy() ways collecting jlmerclusterperm objects tidy data. Julia interface: interacting Julia objects using JuliaConnectoR package. Reproducibility: using Julia RNG make CPA runs reproducible. Asynchronous CPA: running CPA background process using future package.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"compared-to-other-implementations","dir":"Articles","previous_headings":"","what":"Compared to other implementations","title":"Introduction to jlmerclusterperm","text":"R packages cluster-based permutation analysis, clusterperm, eyetrackingR, permuco, permutes. Compared existing implementations, jlmerclusterperm package optimized CPAs based mixed-effects regression models, suitable typical experimental research data multiple, crossed grouping structures (e.g., subjects items). also package interface individual algorithmic steps CPA. Lastly, thanks Julia back-end, bootstrapping fast even mixed models.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/jlmerclusterperm.html","id":"further-readings","dir":"Articles","previous_headings":"","what":"Further readings","title":"Introduction to jlmerclusterperm","text":"original method paper CPA Maris & Oostenveld 2007 recent overview paper Meyer et al. 2021 (CPA EEG research, applies broadly) critical paper Sassenhagen & Draschkow 2019 DOs DON’Ts interpreting reporting CPA results. review paper eyetracking analysis methods Ito & Knoeferle 2022 compares CPA statistical techniques testing difference groups time series data. JOSS paper permuco package (Frossard & Renaud 2021) references therein. Benedikt V. Ehinger’s blog post, includes detailed graphical explainer CPA procedure.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Julia interface","text":"","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"interacting-with-julia-objects","dir":"Articles","previous_headings":"","what":"Interacting with Julia objects","title":"Julia interface","text":"jlmerclusterperm functions collect Julia objects R objects, except jlmer() to_jlmer() return GLM.jl MixedModels.jl fitted model objects. can use functions JuliaConnectoR interact (pointers ) Julia objects: can also call functions Julia packages already installed. example, Effects.jl marginaleffects/emmeans-like features: Note jlmer() to_lmer() just conveniences sanity checking steps cluster-based permutation test, thus used general interface fitting regression models Julia serious analyses.","code":"jmod <- to_jlmer(Reaction ~ Days + (Days | Subject), lme4::sleepstudy) jmod #> #> Variance components: #> Column Variance Std.Dev. Corr. #> Subject (Intercept) 565.51067 23.78047 #> Days 32.68212 5.71683 +0.08 #> Residual 654.94145 25.59182 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 251.405 6.63226 37.91 <1e-99 #> Days 10.4673 1.50224 6.97 <1e-11 #> ────────────────────────────────────────────────── library(JuliaConnectoR) juliaLet(\"x.rePCA\", x = jmod) #> #> (Subject = [0.5406660682947617, 1.0],) juliaCall(\"issingular\", jmod) #> [1] FALSE juliaEval(\"using Effects\") juliaLet(\"effects(x.namedelements, y)\", x = list(Days = 2:5), y = jmod) #> #> 4×5 DataFrame #> Row │ Days Reaction err lower upper #> │ Int64 Float64 Float64 Float64 Float64 #> ─────┼──────────────────────────────────────────── #> 1 │ 2 272.34 7.09427 265.245 279.434 #> 2 │ 3 282.807 7.70544 275.102 290.512 #> 3 │ 4 293.274 8.55557 284.719 301.83 #> 4 │ 5 303.742 9.58128 294.16 313.323"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"julia-global-options","dir":"Articles","previous_headings":"","what":"Julia global options","title":"Julia interface","text":"functions involved cluster-based permutation analysis print progress bars update real time console. function julia_progress() controls whether show progress bar width printed progress bar.","code":"# Set Julia progress options and save old state old_progress_opts <- julia_progress(show = FALSE, width = 30) old_progress_opts #> $show #> [1] TRUE #> #> $width #> [1] 50 julia_progress() #> $show #> [1] FALSE #> #> $width #> [1] 30 # Restored old state julia_progress(old_progress_opts) identical(old_progress_opts, julia_progress()) #> [1] TRUE # The syntax to reset progress options julia_progress(show = TRUE, width = \"auto\") julia_progress()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/julia-interface.html","id":"the-julia-environment","dir":"Articles","previous_headings":"","what":"The Julia environment","title":"Julia interface","text":"following packages loaded Julia environment via jlmerclusterperm_setup():","code":"cat(readLines(system.file(\"julia/Project.toml\", package = \"jlmerclusterperm\")), sep = \"\\n\") #> [deps] #> DataFrames = \"a93c6f00-e57d-5684-b7b6-d8193f3e46c0\" #> Distributions = \"31c24e10-a181-5473-b8eb-7969acd0382f\" #> GLM = \"38e38edf-8417-5370-95a0-9cbb8c7f171a\" #> JlmerClusterPerm = \"2a59065a-9564-4903-82dd-0e42ce19d0e1\" #> MixedModels = \"ff71e718-51f3-5ec2-a782-8ffcbfa3c316\" #> ProgressMeter = \"92933f4c-e287-5a05-a399-4b506db050ca\" #> Random = \"9a3f8284-a2c9-5f02-9a11-845980a1fd5c\" #> Random123 = \"74087812-796a-5b5d-8853-05524746bad3\" #> StatsBase = \"2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91\" #> StatsModels = \"3eaba693-59b7-5ba5-a881-562e759f1c8d\" #> Suppressor = \"fd094767-a336-5f1f-9728-57cf17d0bbfb\" #> #> [compat] #> DataFrames = \"1.5\" #> GLM = \"1.8.2\" #> MixedModels = \"4.12\" #> ProgressMeter = \"1.7\" #> Random123 = \"1.6\" #> StatsBase = \"0.33\" #> StatsModels = \"0.7\" #> Suppressor = \"0.2.1\" #> julia = \"1.8\""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Reproducibility","text":"Minimal example data: jlmer specification object used showcase:","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) library(dplyr, warn.conflicts = FALSE) library(ggplot2) chickweights <- as_tibble(ChickWeight) %>% mutate(diet = forcats::fct_collapse(Diet, A = c(1, 3), B = c(2, 4))) %>% filter(Time %in% 0:20) ggplot(chickweights, aes(Time, weight, color = diet)) + stat_summary(fun.args = list(mult = 1.96)) + stat_summary(geom = \"line\") + theme_minimal(base_size = 18) jlmer_spec <- make_jlmer_spec( formula = weight ~ diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) jlmer_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + dietB + (1 | Chick) #> Predictors: #> diet: dietB #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight dietB Chick Time #> #> 1 42 0 1 0 #> 2 51 0 1 2 #> 3 59 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"rng-seed-vs--counter","dir":"Articles","previous_headings":"","what":"RNG seed vs. counter","title":"Reproducibility","text":"CPA powered bootstrapping, inherently stochastic procedure. Reproducibility guaranteed clusterpermute() calls ran seed counter values. properties Julia random number generator used jlmerclusterperm refer two slightly different concepts: seed conceptually equivalent “seed” talked R, set.seed(). seed used RNG’s initialization, locking specific state persists session. jlmerclusterperm, default seed 1 primarily controlled via options(\"jlmerclusterperm.seed\") must set calling jlmerclusterperm_setup(). counter initializes 0 increments value every time RNG consumed produce random number. equivalent concept base R, though ’s handy intuitive way interfacing RNG’s state interactive session. counter can manipulated via functions reset_rng_state(), set_rng_state(), get_rng_state(). rest vignette showcase interacting counter state, useful practice.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"behavior-of-rng-counter","dir":"Articles","previous_headings":"","what":"Behavior of RNG counter","title":"Reproducibility","text":"start fresh session, counter initialized zero. specify seed, ’s one: start simple clusterpermute() 100 simulations. reproducibility profile run seed 1 counter 0: run, counter now incremented 3712: call clusterpermute() now runs new counter value, produces different result. , expected given stochastic nature CPA (thus need run tens thousands simulations reporting results): second run also increments counter state, note ’s exactly amount (possible predict): difference two runs specifically null distribution cluster-mass statistics. 100-simulation samples drawn share underlying (null) distribution, though identical one another (rarely ).","code":"get_rng_state() #> [1] 0 get_rng_seed() #> [1] 1 CPA_1_0 <- clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) CPA_1_0 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 100) #> Mean (SD): 0.724 (5.54) #> Coverage intervals: 95% [-10.847, 12.592] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0198) #> ──────────────────────────────────────────────────────────────────────────────── get_rng_state() #> [1] 3712 CPA_1_3712 <- clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) CPA_1_3712 #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 100) #> Mean (SD): -0.618 (8.51) #> Coverage intervals: 95% [-18.328, 19.076] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0792) #> ──────────────────────────────────────────────────────────────────────────────── get_rng_state() #> [1] 7427 ggplot(mapping = aes(sum_statistic)) + geom_density( aes(fill = \"1\"), alpha = .5, data = tidy(CPA_1_0$null_cluster_dists) ) + geom_density( aes(fill = \"3712\"), alpha = .5, data = tidy(CPA_1_3712$null_cluster_dists) ) + guides(fill = guide_legend(\"Counter\")) + theme_classic()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"manipulating-rng-counter","dir":"Articles","previous_headings":"","what":"Manipulating RNG counter","title":"Reproducibility","text":"Using set_rng_counter(), possible reproduce prior run. example, reproduce second clusterpermute() call, set counter back 3712 set_rng_state(): results subsequent run counter state identical previous run: reproduce first CPA ran counter 0, can use set_rng_state() simply use reset_rng_state() alias set_rng_state(0): behavior can observed specific step simulating null distribution well. function corresponding step permute_timewise_statistics() run 100 simulations step counter zero: Running cluster detection algorithm result using extract_null_cluster_dists() returns null cluster-mass distribution identical CPA_1_0:","code":"set_rng_state(3712) get_rng_state() #> [1] 3712 identical( CPA_1_3712, clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) ) #> [1] TRUE reset_rng_state() get_rng_state() #> [1] 0 identical( CPA_1_0, clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) ) #> [1] TRUE reset_rng_state() get_rng_state() #> [1] 0 null_statistics <- permute_timewise_statistics(jlmer_spec, nsim = 100) get_rng_state() #> [1] 3712 null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 1) null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 100) #> Mean (SD): 0.724 (5.54) #> Coverage intervals: 95% [-10.847, 12.592] #> ──────────────────────────────────────────────────────────────────────────────── identical(null_cluster_dists, CPA_1_0$null_cluster_dists) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"technical-details-and-notes","dir":"Articles","previous_headings":"","what":"Technical details and notes","title":"Reproducibility","text":"Julia back-end uses thread-safe, counter-based RNG called Threefry (specifically, Random123.Threefry2x) comes Random123.jl Julia library adaptation original algorithm developed Salmon, Moraes, Dror, Shaw (2011). RNG initialized “rounds” value 20, also default. Threefry used speed stability across Julia versions. Additionally, noted earlier, RNG initializes explicit seed 1. behavior different case R, random seed chosen user every new session. consequence absence explicit seeding via options(\"jlmerclusterperm.seed\"), first run every new session (.e., runs counter value zero) identical. Given interface RNG counter, many unique advantages using different seed manipulating RNG counter. interface RNG seed also provided via set_rng_seed() get_rng_seed(). Lastly, please aware changing seed R session (e.g., via set.seed()) effect jlmerclusterperm functions powered Julia back-end.","code":"# Restart Julia session jlmerclusterperm_setup(restart = TRUE, verbose = FALSE) # RNG seed is again 1 get_rng_seed() #> [1] 1 # So results of the first CPA of the session (with counter = 0) is the same get_rng_state() #> [1] 0 identical( CPA_1_0, clusterpermute(jlmer_spec, threshold = 1, nsim = 100, progress = FALSE) ) #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/reproducibility.html","id":"miscellaneous","dir":"Articles","previous_headings":"","what":"Miscellaneous","title":"Reproducibility","text":"may wondering whether two CPAs overlap range counter values also share overlapping samples null. actually case practice shuffling CPA implemented: two colored lines represent two 100-simulation bootstrap permutations differ initial RNG counter value 1. point shape two lines look different - one just shifted form another. formal investigations (e.g., simulating power CPA), ’s course best practice generate random seeds. counter convenience balances interactivity reproducibility.","code":"# 100 simulations of timewise statistics ## First CPA using counter at 1 set_rng_state(1) x <- permute_timewise_statistics(jlmer_spec, nsim = 100) ## Second CPA using counter at 2 set_rng_state(2) y <- permute_timewise_statistics(jlmer_spec, nsim = 100) # The null cluster-mass statistics x_null <- extract_null_cluster_dists(x, threshold = 1) x_null_stats <- tidy(x_null)$sum_statistic y_null <- extract_null_cluster_dists(y, threshold = 1) y_null_stats <- tidy(y_null)$sum_statistic # Relationship between two CPAs with similar initial counters matplot(cbind(x_null_stats, y_null_stats), type = \"b\", pch = 1)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Tidying output","text":"Minimal example data: jlmer specification object used showcase:","code":"library(jlmerclusterperm) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) library(dplyr, warn.conflicts = FALSE) library(ggplot2) chickweights <- as_tibble(ChickWeight) %>% mutate(diet = forcats::fct_collapse(Diet, A = c(1, 3), B = c(2, 4))) %>% filter(Time %in% 0:20) ggplot(chickweights, aes(Time, weight, color = diet)) + stat_summary(fun.args = list(mult = 1.96)) + stat_summary(geom = \"line\") + theme_minimal(base_size = 18) jlmer_spec <- make_jlmer_spec( formula = weight ~ diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) jlmer_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + dietB + (1 | Chick) #> Predictors: #> diet: dietB #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight dietB Chick Time #> #> 1 42 0 1 0 #> 2 51 0 1 2 #> 3 59 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"a-tour-of-tidy-methods","dir":"Articles","previous_headings":"","what":"A tour of tidy() methods","title":"Tidying output","text":"non- objects returned jlmerclusterperm functions tidy() method return data frame representation object.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"timewise_statistics-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":"Functions compute_timewise_statistics() permute_timewise_statistics() return object class , array statistics. Note isn’t stylized print() method class. Instead, recommended use tidy() inspect results data frame statistic represented row. case compute_timewise_statistics(), output matrix (2D array) predictor time. dimension becomes column tidy() data: case compute_timewise_statistics(), output 3D array simulation, time, predictor. , dimension becomes column tidy() data:","code":"empirical_statistics <- compute_timewise_statistics(jlmer_spec) class(empirical_statistics) #> [1] \"timewise_statistics\" dim(empirical_statistics) # predictor x time #> [1] 1 11 tidy(empirical_statistics) #> # A tibble: 11 × 3 #> predictor time statistic #> #> 1 dietB 0 -1.09 #> 2 dietB 2 2.29 #> 3 dietB 4 3.14 #> 4 dietB 6 4.01 #> 5 dietB 8 2.90 #> 6 dietB 10 2.44 #> 7 dietB 12 2.18 #> 8 dietB 14 1.26 #> 9 dietB 16 0.673 #> 10 dietB 18 0.532 #> 11 dietB 20 0.809 null_statistics <- permute_timewise_statistics(jlmer_spec, nsim = 1000) class(null_statistics) #> [1] \"timewise_statistics\" dim(null_statistics) # simulation x time x predictor #> [1] 1000 11 1 tidy(null_statistics) #> # A tibble: 11,000 × 4 #> predictor time statistic sim #> #> 1 dietB 0 -0.309 0001 #> 2 dietB 0 2.10 0002 #> 3 dietB 0 -2.83 0003 #> 4 dietB 0 -0.568 0004 #> 5 dietB 0 2.10 0005 #> 6 dietB 0 -1.09 0006 #> 7 dietB 0 -0.830 0007 #> 8 dietB 0 -0.0515 0008 #> 9 dietB 0 -0.0515 0009 #> 10 dietB 0 0.206 0010 #> # ℹ 10,990 more rows"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"empirical_clusters-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_empirical_clusters() print method defined, makes easier eye: tidy() data frame representation also contains information like length, number time points cluster spans. id variable uniquely identifies cluster within predictor.","code":"empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 1) class(empirical_clusters) #> [1] \"empirical_clusters\" empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 #> ──────────────────────────────────────────────────────────────────────────────── tidy(empirical_clusters) #> # A tibble: 1 × 6 #> predictor id start end length sum_statistic #> #> 1 dietB 1 2 14 7 18.2"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"null_cluster_dists-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_null_cluster_dists() also print method defined, displays summary statistics: null distribution constructed largest cluster-mass statistic observed bootstrapped permutations original data. words, simulation contributes one sample null. tidy() representation, simulation predictor represents row. sum_statistic column represents size largest cluster simulation. clusters found, column zero values start, end, length NA. output tidy(null_statistics) tidy(null_cluster_dists) form relational data. example, can use left_join() find time wise statistics permuted data extreme cluster-mass statistic: Note sum_statistic simply sum statistic values span largest cluster:","code":"null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 1) class(null_cluster_dists) #> [1] \"null_cluster_dists\" null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.014 (7.11) #> Coverage intervals: 95% [-15.360, 15.902] #> ──────────────────────────────────────────────────────────────────────────────── tidy(null_cluster_dists) #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 8 20 7 11.1 0001 #> 2 dietB 6 10 3 3.75 0002 #> 3 dietB 18 20 2 2.07 0003 #> 4 dietB 2 10 5 6.00 0004 #> 5 dietB 4 10 4 5.91 0005 #> 6 dietB NA NA NA 0 0006 #> 7 dietB 8 20 7 12.2 0007 #> 8 dietB 2 8 4 5.29 0008 #> 9 dietB NA NA NA 0 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows largest_permuted_cluster <- tidy(null_cluster_dists) %>% slice(which.max(abs(sum_statistic))) %>% left_join(tidy(null_statistics), by = c(\"predictor\", \"sim\")) largest_permuted_cluster #> # A tibble: 11 × 8 #> predictor start end length sum_statistic sim time statistic #> #> 1 dietB 2 20 10 29.4 0319 0 -1.09 #> 2 dietB 2 20 10 29.4 0319 2 1.67 #> 3 dietB 2 20 10 29.4 0319 4 2.56 #> 4 dietB 2 20 10 29.4 0319 6 2.97 #> 5 dietB 2 20 10 29.4 0319 8 3.47 #> 6 dietB 2 20 10 29.4 0319 10 3.54 #> 7 dietB 2 20 10 29.4 0319 12 3.49 #> 8 dietB 2 20 10 29.4 0319 14 3.54 #> 9 dietB 2 20 10 29.4 0319 16 3.05 #> 10 dietB 2 20 10 29.4 0319 18 2.65 #> 11 dietB 2 20 10 29.4 0319 20 2.48 largest_permuted_cluster %>% filter(between(time, start, end)) %>% summarize(sum_statistic = unique(sum_statistic), sum(statistic)) #> # A tibble: 1 × 2 #> sum_statistic `sum(statistic)` #> #> 1 29.4 29.4"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"in-other-contexts","dir":"Articles","previous_headings":"A tour of tidy() methods","what":"In other contexts","title":"Tidying output","text":"output calculate_clusters_pvalues() just object augmented p-values: available, p-values returned column tidy() data: Note p-value cluster 0.027 27 samples null extreme observed: case one-fell-swoop function clusterpermute(), object object returned list: can tidy() elements list lapply() purrr::map():","code":"tested_clusters <- calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) tested_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0270) #> ──────────────────────────────────────────────────────────────────────────────── tidy(tested_clusters) #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.027 tidy(null_cluster_dists) %>% filter(abs(sum_statistic) >= abs(tidy(tested_clusters)$sum_statistic)) #> # A tibble: 27 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 0 20 11 -18.7 0072 #> 2 dietB 0 20 11 -19.4 0114 #> 3 dietB 6 20 8 19.0 0118 #> 4 dietB 0 20 11 -23.6 0137 #> 5 dietB 4 20 9 -21.7 0142 #> 6 dietB 2 20 10 24.7 0268 #> 7 dietB 4 20 9 22.7 0289 #> 8 dietB 2 20 10 29.4 0319 #> 9 dietB 2 20 10 26.4 0325 #> 10 dietB 4 20 9 26.5 0384 #> # ℹ 17 more rows # Timing the entire process this time system.time({ full_test <- clusterpermute(jlmer_spec, threshold = 1, nsim = 1000, progress = FALSE) }) #> user system elapsed #> 0.00 0.01 1.77 full_test #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.066 (7.61) #> Coverage intervals: 95% [-16.959, 17.216] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0420) #> ──────────────────────────────────────────────────────────────────────────────── lapply(full_test, tidy) #> $null_cluster_dists #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 14 20 4 -5.26 0001 #> 2 dietB 14 20 4 9.22 0002 #> 3 dietB 6 20 8 11.8 0003 #> 4 dietB NA NA NA 0 0004 #> 5 dietB NA NA NA 0 0005 #> 6 dietB 16 20 3 -3.69 0006 #> 7 dietB 14 20 4 6.23 0007 #> 8 dietB NA NA NA 0 0008 #> 9 dietB 12 20 5 5.38 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows #> #> $empirical_clusters #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.0420"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying-output.html","id":"tidying-julia-model-objects","dir":"Articles","previous_headings":"","what":"Tidying Julia model objects","title":"Tidying output","text":"Functions jlmer() to_jlmer() return object class , JuliaConnectoR pointer Julia model object: can use tidy() glance() return information Julia model data frame: mimic behavior methods broom/broom.mixed defined (g)lm/(g)lmer models: Note Julia MixedModels.jl defaults REML = FALSE, whereas TRUE default lme4.","code":"jlmer_mod <- to_jlmer(weight ~ diet + (1 | Chick), chickweights) class(jlmer_mod) #> [1] \"jlmer_mod\" \"JuliaStructProxy\" \"JuliaProxy\" jlmer_mod #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 305.1809 17.4694 #> Residual 3854.0809 62.0812 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 108.16 4.76331 22.71 <1e-99 #> dietB 12.2644 7.45319 1.65 0.0999 #> ────────────────────────────────────────────────── tidy(jlmer_mod) #> # A tibble: 4 × 7 #> effect group term estimate std.error statistic p.value #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 3.83e-114 #> 2 fixed NA dietB 12.3 7.45 1.65 9.99e- 2 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA NA glance(jlmer_mod) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 533 4 62.1 -2972. 5952. 5969. 5944. 529 library(lme4) library(broom.mixed) lmer_mod <- lmer(weight ~ diet + (1 | Chick), chickweights, REML = FALSE) tidy(lmer_mod) #> # A tibble: 4 × 6 #> effect group term estimate std.error statistic #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 #> 2 fixed NA dietB 12.3 7.45 1.65 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA glance(lmer_mod) #> # A tibble: 1 × 7 #> nobs sigma logLik AIC BIC deviance df.residual #> #> 1 533 62.1 -2972. 5952. 5969. 5944. 529"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"setup","dir":"Articles","previous_headings":"","what":"Setup","title":"Tidying output","text":"Minimal example data: jlmer specification object used vignette:","code":"library(jlmerclusterperm) system.time(jlmerclusterperm_setup(verbose = FALSE)) #> user system elapsed #> 0.03 0.03 24.11 julia_progress(show = FALSE) library(dplyr, warn.conflicts = FALSE) library(ggplot2) chickweights <- as_tibble(ChickWeight) %>% mutate(diet = forcats::fct_collapse(Diet, A = c(1, 3), B = c(2, 4))) %>% filter(Time %in% 0:20) ggplot(chickweights, aes(Time, weight, color = diet)) + stat_summary(fun.args = list(mult = 1.96)) + stat_summary(geom = \"line\") + theme_minimal(base_size = 18) jlmer_spec <- make_jlmer_spec( formula = weight ~ diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) jlmer_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + dietB + (1 | Chick) #> Predictors: #> diet: dietB #> Specials: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight dietB Chick Time #> #> 1 42 0 1 0 #> 2 51 0 1 2 #> 3 59 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"a-tour-of-tidy-methods","dir":"Articles","previous_headings":"","what":"A tour of tidy() methods","title":"Tidying output","text":"non- objects returned jlmerclusterperm functions tidy() method return data frame representation object.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"timewise_statistics-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":"Functions compute_timewise_statistics() permute_timewise_statistics() return object class , array statistics. Note isn’t stylized print() method class. Instead, recommended use tidy() inspect results data frame statistic represented row. case compute_timewise_statistics(), output matrix (2D array) predictor time. dimension becomes column tidy() data: case compute_timewise_statistics(), output 3D array simulation, time, predictor. , dimension becomes column tidy() data:","code":"empirical_statistics <- compute_timewise_statistics(jlmer_spec) class(empirical_statistics) #> [1] \"timewise_statistics\" dim(empirical_statistics) # predictor x time #> [1] 1 11 tidy(empirical_statistics) #> # A tibble: 11 × 3 #> predictor time statistic #> #> 1 dietB 0 -1.09 #> 2 dietB 2 2.29 #> 3 dietB 4 3.14 #> 4 dietB 6 4.01 #> 5 dietB 8 2.90 #> 6 dietB 10 2.44 #> 7 dietB 12 2.18 #> 8 dietB 14 1.26 #> 9 dietB 16 0.673 #> 10 dietB 18 0.532 #> 11 dietB 20 0.809 null_statistics <- permute_timewise_statistics(jlmer_spec, nsim = 1000) class(null_statistics) #> [1] \"timewise_statistics\" dim(null_statistics) # simulation x time x predictor #> [1] 1000 11 1 tidy(null_statistics) #> # A tibble: 11,000 × 4 #> predictor time statistic sim #> #> 1 dietB 0 -0.309 0001 #> 2 dietB 0 2.10 0002 #> 3 dietB 0 -2.83 0003 #> 4 dietB 0 -0.568 0004 #> 5 dietB 0 2.10 0005 #> 6 dietB 0 -1.09 0006 #> 7 dietB 0 -0.830 0007 #> 8 dietB 0 -0.0515 0008 #> 9 dietB 0 -0.0515 0009 #> 10 dietB 0 0.206 0010 #> # ℹ 10,990 more rows"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"empirical_clusters-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_empirical_clusters() print method defined, makes easier eye: tidy() data frame representation also contains information like length, number time points cluster spans. id variable uniquely identifies cluster within predictor.","code":"empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 1) class(empirical_clusters) #> [1] \"empirical_clusters\" empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 #> ──────────────────────────────────────────────────────────────────────────────── tidy(empirical_clusters) #> # A tibble: 1 × 6 #> predictor id start end length sum_statistic #> #> 1 dietB 1 2 14 7 18.2"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"null_cluster_dists-objects","dir":"Articles","previous_headings":"A tour of tidy() methods","what":" objects","title":"Tidying output","text":" object returned extract_null_cluster_dists() also print method defined, displays summary statistics: null distribution constructed largest cluster-mass statistic observed bootstrapped permutations original data. words, simulation contributes one sample null. tidy() representation, simulation predictor represents row. sum_statistic column represents size largest cluster simulation. clusters found, column zero values start, end, length NA. output tidy(null_statistics) tidy(null_cluster_dists) form relational data. example, can use left_join() find time wise statistics permuted data extreme cluster-mass statistic: Note sum_statistic simply sum statistic values span largest cluster:","code":"null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 1) class(null_cluster_dists) #> [1] \"null_cluster_dists\" null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.014 (7.11) #> Coverage intervals: 95% [-15.360, 15.902] #> ──────────────────────────────────────────────────────────────────────────────── tidy(null_cluster_dists) #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 8 20 7 11.1 0001 #> 2 dietB 6 10 3 3.75 0002 #> 3 dietB 18 20 2 2.07 0003 #> 4 dietB 2 10 5 6.00 0004 #> 5 dietB 4 10 4 5.91 0005 #> 6 dietB NA NA NA 0 0006 #> 7 dietB 8 20 7 12.2 0007 #> 8 dietB 2 8 4 5.29 0008 #> 9 dietB NA NA NA 0 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows largest_permuted_cluster <- tidy(null_cluster_dists) %>% slice(which.max(abs(sum_statistic))) %>% left_join(tidy(null_statistics), by = c(\"predictor\", \"sim\")) largest_permuted_cluster #> # A tibble: 11 × 8 #> predictor start end length sum_statistic sim time statistic #> #> 1 dietB 2 20 10 29.4 0319 0 -1.09 #> 2 dietB 2 20 10 29.4 0319 2 1.67 #> 3 dietB 2 20 10 29.4 0319 4 2.56 #> 4 dietB 2 20 10 29.4 0319 6 2.97 #> 5 dietB 2 20 10 29.4 0319 8 3.47 #> 6 dietB 2 20 10 29.4 0319 10 3.54 #> 7 dietB 2 20 10 29.4 0319 12 3.49 #> 8 dietB 2 20 10 29.4 0319 14 3.54 #> 9 dietB 2 20 10 29.4 0319 16 3.05 #> 10 dietB 2 20 10 29.4 0319 18 2.65 #> 11 dietB 2 20 10 29.4 0319 20 2.48 largest_permuted_cluster %>% filter(between(time, start, end)) %>% summarize(sum_statistic = unique(sum_statistic), sum(statistic)) #> # A tibble: 1 × 2 #> sum_statistic `sum(statistic)` #> #> 1 29.4 29.4"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"in-other-contexts","dir":"Articles","previous_headings":"A tour of tidy() methods","what":"In other contexts","title":"Tidying output","text":"output calculate_clusters_pvalues() just object augmented p-values: available, p-values returned column tidy() data: Note p-value cluster 0.027 27 samples null extreme observed: case one-fell-swoop function clusterpermute(), object object returned list: can tidy() elements list lapply() purrr::map():","code":"tested_clusters <- calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) tested_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0270) #> ──────────────────────────────────────────────────────────────────────────────── tidy(tested_clusters) #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.027 tidy(null_cluster_dists) %>% filter(abs(sum_statistic) >= abs(tidy(tested_clusters)$sum_statistic)) #> # A tibble: 27 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 0 20 11 -18.7 0072 #> 2 dietB 0 20 11 -19.4 0114 #> 3 dietB 6 20 8 19.0 0118 #> 4 dietB 0 20 11 -23.6 0137 #> 5 dietB 4 20 9 -21.7 0142 #> 6 dietB 2 20 10 24.7 0268 #> 7 dietB 4 20 9 22.7 0289 #> 8 dietB 2 20 10 29.4 0319 #> 9 dietB 2 20 10 26.4 0325 #> 10 dietB 4 20 9 26.5 0384 #> # ℹ 17 more rows # Timing it for fun this time system.time({ full_test <- clusterpermute(jlmer_spec, threshold = 1, nsim = 1000, progress = FALSE) }) #> user system elapsed #> 0.00 0.01 6.53 full_test$null_cluster_dists #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> dietB (n = 1000) #> Mean (SD): 0.066 (7.61) #> Coverage intervals: 95% [-16.959, 17.216] #> ──────────────────────────────────────────────────────────────────────────────── full_test$empirical_clusters #> ── Empirical clusters (t > 1) ────────────────────────── ── #> dietB #> [2, 14]: 18.231 (p=0.0420) #> ──────────────────────────────────────────────────────────────────────────────── lapply(full_test, tidy) #> $null_cluster_dists #> # A tibble: 1,000 × 6 #> predictor start end length sum_statistic sim #> #> 1 dietB 14 20 4 -5.26 0001 #> 2 dietB 14 20 4 9.22 0002 #> 3 dietB 6 20 8 11.8 0003 #> 4 dietB NA NA NA 0 0004 #> 5 dietB NA NA NA 0 0005 #> 6 dietB 16 20 3 -3.69 0006 #> 7 dietB 14 20 4 6.23 0007 #> 8 dietB NA NA NA 0 0008 #> 9 dietB 12 20 5 5.38 0009 #> 10 dietB NA NA NA 0 0010 #> # ℹ 990 more rows #> #> $empirical_clusters #> # A tibble: 1 × 7 #> predictor id start end length sum_statistic pvalue #> #> 1 dietB 1 2 14 7 18.2 0.0420"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/articles/tidying_output.html","id":"tidying-julia-model-objects","dir":"Articles","previous_headings":"","what":"Tidying Julia model objects","title":"Tidying output","text":"Functions jlmer() to_jlmer() return object class , JuliaConnectoR pointer Julia model object: can use tidy() glance() return information Julia model data frame: mimic behavior methods broom/broom.mixed defined (g)lm/(g)lmer models: Note Julia MixedModels.jl defaults REML = FALSE, whereas TRUE default lme4.","code":"jlmer_mod <- to_jlmer(weight ~ diet + (1 | Chick), chickweights) class(jlmer_mod) #> [1] \"jlmer_mod\" \"JuliaStructProxy\" \"JuliaProxy\" jlmer_mod #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 305.1809 17.4694 #> Residual 3854.0809 62.0812 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 108.16 4.76331 22.71 <1e-99 #> dietB 12.2644 7.45319 1.65 0.0999 #> ────────────────────────────────────────────────── tidy(jlmer_mod) #> # A tibble: 4 × 7 #> effect group term estimate std.error statistic p.value #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 3.83e-114 #> 2 fixed NA dietB 12.3 7.45 1.65 9.99e- 2 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA NA glance(jlmer_mod) #> # A tibble: 1 × 7 #> nobs sigma logLik AIC BIC deviance df.residual #> #> 1 533 62.1 -2972. 5952. 5969. 5944. 529 library(lme4) library(broom.mixed) lmer_mod <- lmer(weight ~ diet + (1 | Chick), chickweights, REML = FALSE) tidy(lmer_mod) #> # A tibble: 4 × 6 #> effect group term estimate std.error statistic #> #> 1 fixed NA (Intercept) 108. 4.76 22.7 #> 2 fixed NA dietB 12.3 7.45 1.65 #> 3 ran_pars Chick sd__(Intercept) 17.5 NA NA #> 4 ran_pars Residual sd__Observation 62.1 NA NA glance(lmer_mod) #> # A tibble: 1 × 7 #> nobs sigma logLik AIC BIC deviance df.residual #> #> 1 533 62.1 -2972. 5952. 5969. 5944. 529"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"June Choe. Author, maintainer, copyright holder.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Choe J (2023). jlmerclusterperm: Cluster-Based Permutation Analysis Densely Sampled Time Data. https://github.com/yjunechoe/jlmerclusterperm, https://yjunechoe.github.io/jlmerclusterperm/.","code":"@Manual{, title = {jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data}, author = {June Choe}, year = {2023}, note = {https://github.com/yjunechoe/jlmerclusterperm, https://yjunechoe.github.io/jlmerclusterperm/}, }"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"our-pledge","dir":"","previous_headings":"","what":"Our Pledge","title":"Contributor Covenant Code of Conduct","text":"members, contributors, leaders pledge make participation community harassment-free experience everyone, regardless age, body size, visible invisible disability, ethnicity, sex characteristics, gender identity expression, level experience, education, socio-economic status, nationality, personal appearance, race, caste, color, religion, sexual identity orientation. pledge act interact ways contribute open, welcoming, diverse, inclusive, healthy community.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"our-standards","dir":"","previous_headings":"","what":"Our Standards","title":"Contributor Covenant Code of Conduct","text":"Examples behavior contributes positive environment community include: Demonstrating empathy kindness toward people respectful differing opinions, viewpoints, experiences Giving gracefully accepting constructive feedback Accepting responsibility apologizing affected mistakes, learning experience Focusing best just us individuals, overall community Examples unacceptable behavior include: use sexualized language imagery, sexual attention advances kind Trolling, insulting derogatory comments, personal political attacks Public private harassment Publishing others’ private information, physical email address, without explicit permission conduct reasonably considered inappropriate professional setting","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"enforcement-responsibilities","dir":"","previous_headings":"","what":"Enforcement Responsibilities","title":"Contributor Covenant Code of Conduct","text":"Community leaders responsible clarifying enforcing standards acceptable behavior take appropriate fair corrective action response behavior deem inappropriate, threatening, offensive, harmful. Community leaders right responsibility remove, edit, reject comments, commits, code, wiki edits, issues, contributions aligned Code Conduct, communicate reasons moderation decisions appropriate.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"scope","dir":"","previous_headings":"","what":"Scope","title":"Contributor Covenant Code of Conduct","text":"Code Conduct applies within community spaces, also applies individual officially representing community public spaces. Examples representing community include using official e-mail address, posting via official social media account, acting appointed representative online offline event.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"enforcement","dir":"","previous_headings":"","what":"Enforcement","title":"Contributor Covenant Code of Conduct","text":"Instances abusive, harassing, otherwise unacceptable behavior may reported community leaders responsible enforcement codeofconduct@rstudio.com. complaints reviewed investigated promptly fairly. community leaders obligated respect privacy security reporter incident.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"enforcement-guidelines","dir":"","previous_headings":"","what":"Enforcement Guidelines","title":"Contributor Covenant Code of Conduct","text":"Community leaders follow Community Impact Guidelines determining consequences action deem violation Code Conduct:","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_1-correction","dir":"","previous_headings":"Enforcement Guidelines","what":"1. Correction","title":"Contributor Covenant Code of Conduct","text":"Community Impact: Use inappropriate language behavior deemed unprofessional unwelcome community. Consequence: private, written warning community leaders, providing clarity around nature violation explanation behavior inappropriate. public apology may requested.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_2-warning","dir":"","previous_headings":"Enforcement Guidelines","what":"2. Warning","title":"Contributor Covenant Code of Conduct","text":"Community Impact: violation single incident series actions. Consequence: warning consequences continued behavior. interaction people involved, including unsolicited interaction enforcing Code Conduct, specified period time. includes avoiding interactions community spaces well external channels like social media. Violating terms may lead temporary permanent ban.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_3-temporary-ban","dir":"","previous_headings":"Enforcement Guidelines","what":"3. Temporary Ban","title":"Contributor Covenant Code of Conduct","text":"Community Impact: serious violation community standards, including sustained inappropriate behavior. Consequence: temporary ban sort interaction public communication community specified period time. public private interaction people involved, including unsolicited interaction enforcing Code Conduct, allowed period. Violating terms may lead permanent ban.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"id_4-permanent-ban","dir":"","previous_headings":"Enforcement Guidelines","what":"4. Permanent Ban","title":"Contributor Covenant Code of Conduct","text":"Community Impact: Demonstrating pattern violation community standards, including sustained inappropriate behavior, harassment individual, aggression toward disparagement classes individuals. Consequence: permanent ban sort public interaction within community.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CODE_OF_CONDUCT.html","id":"attribution","dir":"","previous_headings":"","what":"Attribution","title":"Contributor Covenant Code of Conduct","text":"Code Conduct adapted Contributor Covenant, version 2.1, available https://www.contributor-covenant.org/version/2/1/code_of_conduct.html. Community Impact Guidelines inspired [Mozilla’s code conduct enforcement ladder][https://github.com/mozilla/inclusion]. answers common questions code conduct, see FAQ https://www.contributor-covenant.org/faq. Translations available https://www.contributor-covenant.org/translations.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":null,"dir":"","previous_headings":"","what":"Contributing to jlmerclusterperm","title":"Contributing to jlmerclusterperm","text":"outlines propose change jlmerclusterperm.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"fixing-typos","dir":"","previous_headings":"","what":"Fixing typos","title":"Contributing to jlmerclusterperm","text":"can fix typos, spelling mistakes, grammatical errors documentation directly using GitHub web interface, long changes made source file. generally means ’ll need edit roxygen2 comments .R, .Rd file. can find .R file generates .Rd reading comment first line.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"bigger-changes","dir":"","previous_headings":"","what":"Bigger changes","title":"Contributing to jlmerclusterperm","text":"want make bigger change, ’s good idea first file issue make sure someone team agrees ’s needed. ’ve found bug, please file issue illustrates bug minimal reprex (also help write unit test, needed).","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"pull-request-process","dir":"","previous_headings":"Bigger changes","what":"Pull request process","title":"Contributing to jlmerclusterperm","text":"Fork package clone onto computer. haven’t done , recommend using usethis::create_from_github(\"yjunechoe/jlmerclusterperm\", fork = TRUE). Install development dependencies devtools::install_dev_deps(), make sure package passes R CMD check running devtools::check(). R CMD check doesn’t pass cleanly, ’s good idea ask help continuing. Create Git branch pull request (PR). recommend using usethis::pr_init(\"brief-description--change\"). Make changes, commit git, create PR running usethis::pr_push(), following prompts browser. title PR briefly describe change. body PR contain Fixes #issue-number. user-facing changes, add bullet top NEWS.md (.e. just first header). Follow style described https://style.tidyverse.org/news.html.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"code-style","dir":"","previous_headings":"Bigger changes","what":"Code style","title":"Contributing to jlmerclusterperm","text":"New code follow tidyverse style guide. can use styler package apply styles, please don’t restyle code nothing PR. use roxygen2, Markdown syntax, documentation. use testthat unit tests. Contributions test cases included easier accept.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/CONTRIBUTING.html","id":"code-of-conduct","dir":"","previous_headings":"","what":"Code of Conduct","title":"Contributing to jlmerclusterperm","text":"Please note jlmerclusterperm project released Contributor Code Conduct. contributing project agree abide terms.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"jlmerclusterperm-","dir":"","previous_headings":"","what":"jlmerclusterperm","title":"jlmerclusterperm","text":"Julia GLM.jl MixedModels.jl based implementation cluster-based permutation test time series data, powered JuliaConnectoR.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"installation-and-usage","dir":"","previous_headings":"","what":"Installation and usage","title":"jlmerclusterperm","text":"Install released version jlmerclusterperm CRAN: install development version GitHub : Using jlmerclusterperm requires prior installation Julia programming language, can downloaded either official website using command line utility juliaup. Julia version >=1.8 required 1.9 preferred substantial speed improvements. using functions jlmerclusterperm, initial setup required via calling jlmerclusterperm_setup(). first call system install necessary dependencies (happens takes around 10-15 minutes). Subsequent calls jlmerclusterperm_setup() incur small overhead around 30 seconds, plus slight delays first-time function calls. pay front start-warm-costs get blazingly-fast functions package. See Get Started page package website background tutorials.","code":"install.packages(\"jlmerclusterperm\") # install.packages(\"remotes\") remotes::install_github(\"yjunechoe/jlmerclusterperm\") # Both lines must be run at the start of each new session library(jlmerclusterperm) jlmerclusterperm_setup()"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"wholesale-cpa-with-clusterpermute","dir":"","previous_headings":"Quick tour of package functionalities","what":"Wholesale CPA with clusterpermute()","title":"jlmerclusterperm","text":"time series data: Preparing specification object make_jlmer_spec(): Cluster-based permutation test clusterpermute(): Including random effects:","code":"chickweights <- ChickWeight chickweights$Time <- as.integer(factor(chickweights$Time)) matplot( tapply(chickweights$weight, chickweights[c(\"Time\", \"Diet\")], mean), type = \"b\", lwd = 3, ylab = \"Weight\", xlab = \"Time\" ) chickweights_spec <- make_jlmer_spec( formula = weight ~ 1 + Diet, data = chickweights, subject = \"Chick\", time = \"Time\" ) chickweights_spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> weight Diet2 Diet3 Diet4 Chick Time #> 1 42 0 0 0 1 1 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 3 #> [ reached 'max' / getOption(\"max.print\") -- omitted 575 rows ] #> ──────────────────────────────────────────────────────────────────────────────── set_rng_state(123L) clusterpermute( chickweights_spec, threshold = 2.5, nsim = 100, progress = FALSE ) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2.5) ──────────── ── #> Diet2 (n = 100) #> Mean (SD): -0.039 (1.89) #> Coverage intervals: 95% [-2.862, 0.000] #> Diet3 (n = 100) #> Mean (SD): -0.129 (2.02) #> Coverage intervals: 95% [0.000, 0.000] #> Diet4 (n = 100) #> Mean (SD): 0.296 (3.21) #> Coverage intervals: 95% [0.000, 5.797] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.121 (p=0.0495) #> Diet3 #> [3, 12]: 35.769 (p=0.0099) #> Diet4 #> [2, 8]: 32.442 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── chickweights_re_spec <- make_jlmer_spec( formula = weight ~ 1 + Diet + (1 | Chick), data = chickweights, subject = \"Chick\", time = \"Time\" ) set_rng_state(123L) clusterpermute( chickweights_re_spec, threshold = 2.5, nsim = 100, progress = FALSE )$empirical_clusters #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.387 (p=0.0594) #> Diet3 #> [2, 12]: 39.919 (p=0.0099) #> Diet4 #> [2, 8]: 33.853 (p=0.0099) #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"piecemeal-approach-to-cpa","dir":"","previous_headings":"Quick tour of package functionalities","what":"Piecemeal approach to CPA","title":"jlmerclusterperm","text":"Computing time-wise statistics observed data: Identifying empirical clusters: Simulating null distribution: Significance testing cluster-mass statistic: Iterating range threshold values:","code":"empirical_statistics <- compute_timewise_statistics(chickweights_spec) matplot(t(empirical_statistics), type = \"b\", pch = 1, lwd = 3, ylab = \"t-statistic\") abline(h = 2.5, lty = 3) empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2.5) empirical_clusters #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.121 #> Diet3 #> [3, 12]: 35.769 #> Diet4 #> [2, 8]: 32.442 #> ──────────────────────────────────────────────────────────────────────────────── set_rng_state(123L) null_statistics <- permute_timewise_statistics(chickweights_spec, nsim = 100) null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2.5) null_cluster_dists #> ── Null cluster-mass distribution (t > 2.5) ──────────── ── #> Diet2 (n = 100) #> Mean (SD): -0.039 (1.89) #> Coverage intervals: 95% [-2.862, 0.000] #> Diet3 (n = 100) #> Mean (SD): -0.129 (2.02) #> Coverage intervals: 95% [0.000, 0.000] #> Diet4 (n = 100) #> Mean (SD): 0.296 (3.21) #> Coverage intervals: 95% [0.000, 5.797] #> ──────────────────────────────────────────────────────────────────────────────── calculate_clusters_pvalues(empirical_clusters, null_cluster_dists, add1 = TRUE) #> ── Empirical clusters (t > 2.5) ──────────────────────── ── #> Diet2 #> [3, 4]: 6.121 (p=0.0495) #> Diet3 #> [3, 12]: 35.769 (p=0.0099) #> Diet4 #> [2, 8]: 32.442 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── walk_threshold_steps(empirical_statistics, null_statistics, steps = c(2, 2.5, 3)) #> threshold predictor id start end length sum_statistic pvalue #> 1 2.0 Diet2 1 3 5 3 8.496376 0.07920792 #> 2 2.0 Diet3 1 2 12 11 38.216035 0.00990099 #> 3 2.0 Diet4 1 2 12 11 41.651468 0.00990099 #> 4 2.5 Diet2 1 3 4 2 6.121141 0.04950495 #> 5 2.5 Diet3 1 3 12 10 35.768957 0.00990099 #> 6 2.5 Diet4 1 2 8 7 32.442352 0.00990099 #> 31 3.0 Diet3 1 3 5 3 12.719231 0.00990099 #> 21 3.0 Diet3 2 9 12 4 14.037622 0.00990099 #> 41 3.0 Diet4 1 2 7 6 29.659402 0.00990099"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"acknowledgments","dir":"","previous_headings":"","what":"Acknowledgments","title":"jlmerclusterperm","text":"paper Maris & Oostenveld (2007) originally proposed cluster-based permutation analysis. JuliaConnectoR package powering R interface Julia. Julia packages GLM.jl MixedModels.jl fast implementations (mixed effects) regression models. Existing implementations CPA R (permuco, permutes, etc.) whose designs inspired CPA interface jlmerclusterperm.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/index.html","id":"citations","dir":"","previous_headings":"","what":"Citations","title":"jlmerclusterperm","text":"use jlmerclusterperm cluster-based permutation test mixed-effects models research, please cite one () following see fit. cite jlmerclusterperm: Choe, J. (2023). jlmerclusterperm: Cluster-Based Permutation Analysis Densely Sampled Time Data. R package version 1.0.4. https://cran.r-project.org/package=jlmerclusterperm. cite cluster-based permutation test: Maris, E., & Oostenveld, R. (2007). Nonparametric statistical testing EEG- MEG-data. Journal Neuroscience Methods, 164, 177–190. doi: 10.1016/j.jneumeth.2007.03.024. cite Julia programming language: Bezanson, J., Edelman, ., Karpinski, S., & Shah, V. B. (2017). Julia: Fresh Approach Numerical Computing. SIAM Review, 59(1), 65–98. doi: 10.1137/141000671. cite GLM.jl MixedModels.jl Julia libraries, consult Zenodo pages: GLM: https://doi.org/10.5281/zenodo.3376013 MixedModels: https://zenodo.org/badge/latestdoi/9106942","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/LICENSE.html","id":null,"dir":"","previous_headings":"","what":"MIT License","title":"MIT License","text":"Copyright (c) 2023 June Choe Permission hereby granted, free charge, person obtaining copy software associated documentation files (“Software”), deal Software without restriction, including without limitation rights use, copy, modify, merge, publish, distribute, sublicense, /sell copies Software, permit persons Software furnished , subject following conditions: copyright notice permission notice shall included copies substantial portions Software. SOFTWARE PROVIDED “”, WITHOUT WARRANTY KIND, EXPRESS IMPLIED, INCLUDING LIMITED WARRANTIES MERCHANTABILITY, FITNESS PARTICULAR PURPOSE NONINFRINGEMENT. EVENT SHALL AUTHORS COPYRIGHT HOLDERS LIABLE CLAIM, DAMAGES LIABILITY, WHETHER ACTION CONTRACT, TORT OTHERWISE, ARISING , CONNECTION SOFTWARE USE DEALINGS SOFTWARE.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"Calculate bootstrapped p-values cluster-mass statistics","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"","code":"calculate_clusters_pvalues( empirical_clusters, null_cluster_dists, add1 = FALSE ) clusters_are_comparable(empirical_clusters, null_cluster_dists, error = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"empirical_clusters empirical_clusters object null_cluster_dists null_cluster_dists object add1 Whether add 1 numerator denominator calculating p-value. Use TRUE effectively count observed statistic part permuted null distribution (recommended larger nsim prior publishing results). error Whether throw error incompatible","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"empirical_clusters object augmented p-values.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/calculate_clusters_pvalues.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate bootstrapped p-values of cluster-mass statistics — calculate_clusters_pvalues","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Make empirical clusters empirical_statistics <- compute_timewise_statistics(spec) empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 #> Diet3 #> [2, 20]: 34.628 #> Diet4 #> [2, 20]: 39.371 #> ──────────────────────────────────────────────────────────────────────────────── # Make null cluster-mass distribution reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) # Significance test the empirical cluster(s) from each predictor against the simulated null calculate_clusters_pvalues(empirical_clusters, null_cluster_dists) #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0200) #> Diet3 #> [2, 20]: 34.628 (p=0.0000) #> Diet4 #> [2, 20]: 39.371 (p=0.0000) #> ──────────────────────────────────────────────────────────────────────────────── # Set `add1 = TRUE` to normalize by adding 1 to numerator and denominator calculate_clusters_pvalues(empirical_clusters, null_cluster_dists, add1 = TRUE) #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # This sequence of procedures is equivalent to `clusterpermute()` reset_rng_state() clusterpermute(spec, threshold = 2, nsim = 100, progress = FALSE) #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # The empirical clusters and the null cluster-mass distribution must be comparable empirical_clusters2 <- extract_empirical_clusters(empirical_statistics, threshold = 3) # For example, below code errors because thresholds are different (2 vs. 3) try( calculate_clusters_pvalues(empirical_clusters2, null_cluster_dists) ) #> Error in clusters_are_comparable(empirical_clusters, null_cluster_dists, : #> Cluster-mass statistics between empirical and null are not comparable. #> ℹ `empirical_clusters` and `null_cluster_dists` must share the same `statistic` #> and `threshold`. #> ✖ threshold: empirical uses 3 but null uses 2. # Check for compatibility with `clusters_are_comparable()` clusters_are_comparable(empirical_clusters, null_cluster_dists) #> [1] TRUE clusters_are_comparable(empirical_clusters2, null_cluster_dists) #> [1] FALSE # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":null,"dir":"Reference","previous_headings":"","what":"Conduct a cluster-based permutation test — clusterpermute","title":"Conduct a cluster-based permutation test — clusterpermute","text":"Conduct cluster-based permutation test","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Conduct a cluster-based permutation test — clusterpermute","text":"","code":"clusterpermute( jlmer_spec, family = c(\"gaussian\", \"binomial\"), statistic = c(\"t\", \"chisq\"), threshold, nsim = 100L, predictors = NULL, binned = FALSE, top_n = Inf, add1 = TRUE, ..., progress = TRUE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Conduct a cluster-based permutation test — clusterpermute","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". statistic Test statistic calculating cluster mass. Can one \"t\" (default) regression model output \"chisq\" likelihood ratio test (takes twice long calculate). threshold threshold value statistic must pass contribute cluster mass. Interpretation differs choice statistic (): statistic = \"t\", threshold t-value (beta/std.err) regression model. statistic = \"chisq\", threshold p-value chi-squared statistics likelihood ratio tests. nsim Number simulations description predictors (Optional) subset predictors test. Defaults NULL tests predictors. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist. top_n many clusters return, order size cluster-mass statistic. Defaults Inf return detected clusters. add1 Whether add 1 numerator denominator calculating p-value. Use TRUE effectively count observed statistic part permuted null distribution (recommended larger nsim prior publishing results). ... Optional arguments passed Julia model fitting. Defaults fast = TRUE (family = \"binomial\") progress = FALSE. progress Defaults TRUE, prints progress step cluster permutation test.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Conduct a cluster-based permutation test — clusterpermute","text":"list null_cluster_dists empirical_clusters p-values","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/clusterpermute.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Conduct a cluster-based permutation test — clusterpermute","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Should minimally provide `threshold` and `nsim`, in addition to the spec object reset_rng_state() CPA <- clusterpermute(spec, threshold = 2, nsim = 100, progress = FALSE) CPA #> $null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── #> #> $empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # CPA is a list of `` and `` objects sapply(CPA, class) #> null_cluster_dists empirical_clusters #> \"null_cluster_dists\" \"empirical_clusters\" # You can extract the individual components for further inspection CPA$null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── CPA$empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 (p=0.0297) #> Diet3 #> [2, 20]: 34.628 (p=0.0099) #> Diet4 #> [2, 20]: 39.371 (p=0.0099) #> ──────────────────────────────────────────────────────────────────────────────── # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":null,"dir":"Reference","previous_headings":"","what":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"Tidiers cluster permutation test objects","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"","code":"# S3 method for timewise_statistics tidy(x, ...) # S3 method for empirical_clusters tidy(x, ...) # S3 method for null_cluster_dists tidy(x, ...)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"x object class , , ... Unused","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"data frame","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/cluster_permutation_tidiers.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tidiers for cluster permutation test objects — cluster_permutation_tidiers","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Method for `` empirical_statistics <- compute_timewise_statistics(spec) class(empirical_statistics) #> [1] \"timewise_statistics\" tidy(empirical_statistics) #> # A tibble: 33 × 3 #> predictor time statistic #> #> 1 Diet2 0 -1.60 #> 2 Diet3 0 -1.37 #> 3 Diet4 0 -0.916 #> 4 Diet2 2 1.67 #> 5 Diet3 2 2.45 #> 6 Diet4 2 3.53 #> 7 Diet2 4 2.60 #> 8 Diet3 4 4.48 #> 9 Diet4 4 6.27 #> 10 Diet2 6 3.52 #> # ℹ 23 more rows reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) class(null_statistics) #> [1] \"timewise_statistics\" tidy(null_statistics) #> # A tibble: 3,300 × 4 #> predictor time statistic sim #> #> 1 Diet2 0 -0.898 001 #> 2 Diet2 0 1.40 002 #> 3 Diet2 0 -1.44 003 #> 4 Diet2 0 -0.449 004 #> 5 Diet2 0 1.38 005 #> 6 Diet2 0 -0.692 006 #> 7 Diet2 0 0.113 007 #> 8 Diet2 0 0.334 008 #> 9 Diet2 0 0.558 009 #> 10 Diet2 0 -0.449 010 #> # ℹ 3,290 more rows # Method for `` empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) class(empirical_clusters) #> [1] \"empirical_clusters\" tidy(empirical_clusters) #> # A tibble: 3 × 6 #> predictor id start end length sum_statistic #> #> 1 Diet2 1 4 8 3 8.50 #> 2 Diet3 1 2 20 10 34.6 #> 3 Diet4 1 2 20 10 39.4 # Method for `` null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) class(null_cluster_dists) #> [1] \"null_cluster_dists\" tidy(null_cluster_dists) #> # A tibble: 300 × 6 #> predictor start end length sum_statistic sim #> #> 1 Diet2 10 20 6 14.8 001 #> 2 Diet2 NA NA NA 0 002 #> 3 Diet2 NA NA NA 0 003 #> 4 Diet2 NA NA NA 0 004 #> 5 Diet2 NA NA NA 0 005 #> 6 Diet2 NA NA NA 0 006 #> 7 Diet2 NA NA NA 0 007 #> 8 Diet2 NA NA NA 0 008 #> 9 Diet2 NA NA NA 0 009 #> 10 Diet2 NA NA NA 0 010 #> # ℹ 290 more rows # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"Fit Julia regression models time point time series data","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"","code":"compute_timewise_statistics( jlmer_spec, family = c(\"gaussian\", \"binomial\"), statistic = c(\"t\", \"chisq\"), ... )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". statistic Test statistic calculating cluster mass. Can one \"t\" (default) regression model output \"chisq\" likelihood ratio test (takes twice long calculate). ... Optional arguments passed Julia model fitting. Defaults fast = TRUE (family = \"binomial\") progress = FALSE.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"predictor--time matrix cluster statistics, class timewise_statistics.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/compute_timewise_statistics.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit Julia regression models to each time point of a time series data — compute_timewise_statistics","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Predictor x Time matrix of t-statistics from regression output empirical_statistics <- compute_timewise_statistics(spec) round(empirical_statistics, 2) #> Time #> Predictor 0 2 4 6 8 10 12 14 16 18 20 #> Diet2 -1.60 1.67 2.60 3.52 2.38 1.91 1.94 1.34 1.16 1.40 1.51 #> Diet3 -1.37 2.45 4.48 4.54 3.70 2.97 3.06 2.98 3.05 3.60 3.80 #> Diet4 -0.92 3.53 6.27 7.00 5.12 4.07 3.66 2.78 2.16 2.13 2.64 # Collect as dataframe with `tidy()` empirical_statistics_df <- tidy(empirical_statistics) empirical_statistics_df #> # A tibble: 33 × 3 #> predictor time statistic #> #> 1 Diet2 0 -1.60 #> 2 Diet3 0 -1.37 #> 3 Diet4 0 -0.916 #> 4 Diet2 2 1.67 #> 5 Diet3 2 2.45 #> 6 Diet4 2 3.53 #> 7 Diet2 4 2.60 #> 8 Diet3 4 4.48 #> 9 Diet4 4 6.27 #> 10 Diet2 6 3.52 #> # ℹ 23 more rows # Timewise statistics are from regression models fitted to each time point # - Note the identical statistics at `Time == 0` empirical_statistics_df %>% filter(time == 0) #> # A tibble: 3 × 3 #> predictor time statistic #> #> 1 Diet2 0 -1.60 #> 2 Diet3 0 -1.37 #> 3 Diet4 0 -0.916 to_jlmer(weight ~ 1 + Diet, filter(ChickWeight, Time == 0)) %>% tidy() %>% select(term, statistic) #> # A tibble: 4 × 2 #> term statistic #> #> 1 (Intercept) 164. #> 2 Diet2 -1.60 #> 3 Diet3 -1.37 #> 4 Diet4 -0.916 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":null,"dir":"Reference","previous_headings":"","what":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"Detect largest clusters time sequence predictor statistics","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"","code":"extract_empirical_clusters( empirical_statistics, threshold, binned = FALSE, top_n = Inf )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"empirical_statistics predictor--time matrix empirical timewise statistics. threshold threshold value statistic must pass contribute cluster mass. Interpretation differs choice statistic (): statistic = \"t\", threshold t-value (beta/std.err) regression model. statistic = \"chisq\", threshold p-value chi-squared statistics likelihood ratio tests. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist. top_n many clusters return, order size cluster-mass statistic. Defaults Inf return detected clusters.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"empirical_clusters object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_empirical_clusters.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Detect largest clusters from a time sequence of predictor statistics — extract_empirical_clusters","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Empirical clusters are derived from the timewise statistics empirical_statistics <- compute_timewise_statistics(spec) empirical_clusters <- extract_empirical_clusters(empirical_statistics, threshold = 2) empirical_clusters #> ── Empirical clusters (t > 2) ────────────────────────── ── #> Diet2 #> [4, 8]: 8.496 #> Diet3 #> [2, 20]: 34.628 #> Diet4 #> [2, 20]: 39.371 #> ──────────────────────────────────────────────────────────────────────────────── # Collect as dataframe with `tidy()` empirical_clusters_df <- tidy(empirical_clusters) empirical_clusters_df #> # A tibble: 3 × 6 #> predictor id start end length sum_statistic #> #> 1 Diet2 1 4 8 3 8.50 #> 2 Diet3 1 2 20 10 34.6 #> 3 Diet4 1 2 20 10 39.4 # Changing the `threshold` value identifies different clusters extract_empirical_clusters(empirical_statistics, threshold = 1) #> ── Empirical clusters (t > 1) ────────────────────────── ── #> Diet2 #> [2, 20]: 19.427 #> Diet3 #> [2, 20]: 34.628 #> Diet4 #> [2, 20]: 39.371 #> ──────────────────────────────────────────────────────────────────────────────── # A predictor can have zero or multiple clusters associated with it extract_empirical_clusters(empirical_statistics, threshold = 3) #> ── Empirical clusters (t > 3) ────────────────────────── ── #> Diet3 #> [4, 8]: 12.719 #> [16, 20]: 10.449 #> Diet4 #> [2, 12]: 29.659 #> ──────────────────────────────────────────────────────────────────────────────── #> ! No clusters found for Diet2 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":null,"dir":"Reference","previous_headings":"","what":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"Construct null distribution cluster-mass statistics","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"","code":"extract_null_cluster_dists(null_statistics, threshold, binned = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"null_statistics simulation--time--predictor 3D array null (permuted) timewise statistics. threshold threshold value statistic must pass contribute cluster mass. Interpretation differs choice statistic (): statistic = \"t\", threshold t-value (beta/std.err) regression model. statistic = \"chisq\", threshold p-value chi-squared statistics likelihood ratio tests. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"null_cluster_dists object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/extract_null_cluster_dists.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Construct a null distribution of cluster-mass statistics — extract_null_cluster_dists","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Null cluster-mass distributions are derived from the permuted timewise statistics reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) null_cluster_dists <- extract_null_cluster_dists(null_statistics, threshold = 2) null_cluster_dists #> ── Null cluster-mass distribution (t > 2) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.266 (2.13) #> Coverage intervals: 95% [0.000, 6.579] #> Diet3 (n = 100) #> Mean (SD): -0.124 (2.75) #> Coverage intervals: 95% [-4.777, 5.707] #> Diet4 (n = 100) #> Mean (SD): 0.007 (2.50) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── # Collect as dataframe with `tidy()` # - Each simulation contributes one (largest) cluster-mass statistic to the null # - When no clusters are found, the `sum_statistic` value is zero null_cluster_dists_df <- tidy(null_cluster_dists) null_cluster_dists_df #> # A tibble: 300 × 6 #> predictor start end length sum_statistic sim #> #> 1 Diet2 10 20 6 14.8 001 #> 2 Diet2 NA NA NA 0 002 #> 3 Diet2 NA NA NA 0 003 #> 4 Diet2 NA NA NA 0 004 #> 5 Diet2 NA NA NA 0 005 #> 6 Diet2 NA NA NA 0 006 #> 7 Diet2 NA NA NA 0 007 #> 8 Diet2 NA NA NA 0 008 #> 9 Diet2 NA NA NA 0 009 #> 10 Diet2 NA NA NA 0 010 #> # ℹ 290 more rows # Changing the `threshold` value changes the shape of the null extract_null_cluster_dists(null_statistics, threshold = 1) #> ── Null cluster-mass distribution (t > 1) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.565 (5.52) #> Coverage intervals: 95% [-9.681, 12.061] #> Diet3 (n = 100) #> Mean (SD): 0.089 (6.27) #> Coverage intervals: 95% [-15.847, 12.185] #> Diet4 (n = 100) #> Mean (SD): 0.715 (5.23) #> Coverage intervals: 95% [-7.843, 11.728] #> ──────────────────────────────────────────────────────────────────────────────── extract_null_cluster_dists(null_statistics, threshold = 3) #> ── Null cluster-mass distribution (t > 3) ────────────── ── #> Diet2 (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> Diet3 (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> Diet4 (n = 100) #> Mean (SD): 0.000 (0.00) #> Coverage intervals: 95% [0.000, 0.000] #> ──────────────────────────────────────────────────────────────────────────────── # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit a Julia regression model using jlmer specifications — jlmer","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"Fit Julia regression model using jlmer specifications","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"","code":"jlmer(jlmer_spec, family = c(\"gaussian\", \"binomial\"), ..., progress = FALSE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". ... Optional arguments passed Julia model fitting. progress TRUE, prints timing iterations.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"jlmer_mod object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmer.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit a Julia regression model using jlmer specifications — jlmer","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } jlmerclusterperm_setup(verbose = FALSE) # Fitting a regression model with a specification object spec <- make_jlmer_spec(weight ~ 1 + Diet, ChickWeight) jlmer(spec) #> #> ──────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ──────────────────────────────────────────────────────────────────────── #> (Intercept) 102.645 4.67395 21.96 <1e-99 93.4847 111.806 #> Diet2 19.9712 7.86744 2.54 0.0111 4.55132 35.3911 #> Diet3 40.3045 7.86744 5.12 <1e-06 24.8847 55.7244 #> Diet4 32.6173 7.91046 4.12 <1e-04 17.113 48.1215 #> ──────────────────────────────────────────────────────────────────────── # `lm()` equivalent summary(lm(weight ~ 1 + Diet, ChickWeight))$coef #> Estimate Std. Error t value Pr(>|t|) #> (Intercept) 102.64545 4.673954 21.961161 4.712762e-78 #> Diet2 19.97121 7.867437 2.538465 1.139708e-02 #> Diet3 40.30455 7.867437 5.122958 4.113938e-07 #> Diet4 32.61726 7.910461 4.123307 4.286352e-05 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm-package.html","id":null,"dir":"Reference","previous_headings":"","what":"jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data — jlmerclusterperm-package","title":"jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data — jlmerclusterperm-package","text":"implementation fast cluster-based permutation analysis (CPA) densely-sampled time data developed Maris & Oostenveld, 2007 doi:10.1016/j.jneumeth.2007.03.024 . Supports (generalized, mixed-effects) regression models calculation timewise statistics. Provides wholesale piecemeal interface CPA procedure emphasis interpretability diagnostics. Integrates 'Julia' libraries 'MixedModels.jl' 'GLM.jl' performance improvements, additional functionalities interfacing 'Julia' 'R' powered 'JuliaConnectoR' package.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"jlmerclusterperm: Cluster-Based Permutation Analysis for Densely Sampled Time Data — jlmerclusterperm-package","text":"Maintainer: June Choe jchoe001@gmail.com (ORCID) [copyright holder]","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":null,"dir":"Reference","previous_headings":"","what":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"Initial setup jlmerclusterperm package","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"","code":"jlmerclusterperm_setup(..., restart = TRUE, verbose = TRUE)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"... Ignored. restart Whether set fresh Julia session, given one already running. FALSE jlmerclusterperm_setup() already called, nothing happens. verbose Print progress messages Julia console","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"TRUE","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/jlmerclusterperm_setup.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Initial setup for the jlmerclusterperm package — jlmerclusterperm_setup","text":"","code":"# \\donttest{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":null,"dir":"Reference","previous_headings":"","what":"Tidier methods for Julia regression models — julia_model_tidiers","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"Tidier methods Julia regression models","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"","code":"# S3 method for jlmer_mod tidy(x, effects = c(\"var_model\", \"ran_pars\", \"fixed\"), ...) # S3 method for jlmer_mod glance(x, ...)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"x object class jlmer_mod effects One \"var_model\", \"ran_pars\", \"fixed\" ... Unused","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"data frame","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_model_tidiers.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tidier methods for Julia regression models — julia_model_tidiers","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } # Fixed-effects only model mod1 <- to_jlmer(weight ~ 1 + Diet, ChickWeight) tidy(mod1) #> # A tibble: 4 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 103. 4.67 22.0 6.77e-107 #> 2 Diet2 20.0 7.87 2.54 1.11e- 2 #> 3 Diet3 40.3 7.87 5.12 3.01e- 7 #> 4 Diet4 32.6 7.91 4.12 3.73e- 5 glance(mod1) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 578 5 69.3 -3268. 6546. 6568. 2758693. 573 # Mixed model mod2 <- to_jlmer(weight ~ 1 + Diet + (1 | Chick), ChickWeight) tidy(mod2) #> # A tibble: 6 × 7 #> effect group term estimate std.error statistic p.value #> #> 1 fixed NA (Intercept) 102. 5.78 17.6 2.74e-69 #> 2 fixed NA Diet2 20.8 9.79 2.13 3.34e- 2 #> 3 fixed NA Diet3 41.2 9.79 4.20 2.63e- 5 #> 4 fixed NA Diet4 33.3 9.83 3.39 7.08e- 4 #> 5 ran_pars Chick sd__(Intercept) 15.7 NA NA NA #> 6 ran_pars Residual sd__Observation 67.3 NA NA NA glance(mod2) #> # A tibble: 1 × 8 #> nobs df sigma logLik AIC BIC deviance df.residual #> #> 1 578 6 67.3 -3265. 6542. 6568. 6530. 572 # Select which of fixed/random effects to return tidy(mod2, effects = \"fixed\") #> # A tibble: 4 × 5 #> term estimate std.error statistic p.value #> #> 1 (Intercept) 102. 5.78 17.6 2.74e-69 #> 2 Diet2 20.8 9.79 2.13 3.34e- 2 #> 3 Diet3 41.2 9.79 4.20 2.63e- 5 #> 4 Diet4 33.3 9.83 3.39 7.08e- 4 tidy(mod2, effects = \"ran_pars\") #> # A tibble: 2 × 3 #> group term estimate #> #> 1 Chick sd__(Intercept) 15.7 #> 2 Residual sd__Observation 67.3 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":null,"dir":"Reference","previous_headings":"","what":"Set/get options for Julia progress bar — julia_progress","title":"Set/get options for Julia progress bar — julia_progress","text":"Set/get options Julia progress bar","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Set/get options for Julia progress bar — julia_progress","text":"","code":"julia_progress(show, width)"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Set/get options for Julia progress bar — julia_progress","text":"show Whether show progress bar. may also pass list \"show\" \"width\". width Width progress bar. \"auto\", adjusts progress bar width fit console.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Set/get options for Julia progress bar — julia_progress","text":"Previous values show width","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_progress.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Set/get options for Julia progress bar — julia_progress","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) # } # Show current progress options julia_progress() #> $show #> [1] TRUE #> #> $width #> [1] 36 #> # Set options and save previous options old_progress_opts <- julia_progress(show = FALSE, width = 100) julia_progress() #> $show #> [1] FALSE #> #> $width #> [1] 100 #> # Restoring progress settings by passing a list of old options old_progress_opts #> $show #> [1] TRUE #> #> $width #> [1] 36 #> julia_progress(old_progress_opts) identical(julia_progress(), old_progress_opts) #> [1] TRUE # Alternatively, reset to default settings using this syntax: julia_progress(show = TRUE, width = \"auto\") # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":null,"dir":"Reference","previous_headings":"","what":"Interface to the Julia RNG — julia_rng","title":"Interface to the Julia RNG — julia_rng","text":"Interface Julia RNG","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Interface to the Julia RNG — julia_rng","text":"","code":"set_rng_state(i) reset_rng_state() get_rng_state() set_rng_seed(seed) get_rng_seed()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Interface to the Julia RNG — julia_rng","text":"Counter number seed Seed","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Interface to the Julia RNG — julia_rng","text":"current seed counter","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_rng.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Interface to the Julia RNG — julia_rng","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } # RNG initializes to seed=1 counter=0 get_rng_seed() #> [1] 1 get_rng_state() #> [1] 0 # setter/getter for RNG counter set_rng_state(123) get_rng_state() #> [1] 123 # setter/getter for RNG seed set_rng_seed(2) get_rng_seed() #> [1] 2 # restore to initial setting (seed=1, counter=0) set_rng_seed(1) set_rng_state(0) # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":null,"dir":"Reference","previous_headings":"","what":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"Check Julia setup requirements jlmerclusterperm","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"","code":"julia_setup_ok()"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"Boolean","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/julia_setup_ok.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Check Julia setup requirements for jlmerclusterperm — julia_setup_ok","text":"","code":"julia_setup_ok() #> [1] TRUE"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":null,"dir":"Reference","previous_headings":"","what":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"Create specifications object fitting regression models Julia","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"","code":"make_jlmer_spec( formula, data, subject = NULL, trial = NULL, time = NULL, drop_terms = NULL, ... )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"formula Model formula R syntax data data frame subject Column subjects data. trial Column trials data. Must uniquely identify time series within subject (example, column items counterbalanced design subject sees exactly one item). time Column time data. drop_terms (Optional) terms drop reconstructed model formula ... Unused, extensibility.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"object class jlmer_spec.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/make_jlmer_spec.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create a specifications object for fitting regression models in Julia — make_jlmer_spec","text":"","code":"# Bare specification object (minimal spec for fitting a global model) spec <- make_jlmer_spec(weight ~ 1 + Diet, ChickWeight) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Data: #> # A tibble: 578 × 4 #> weight Diet2 Diet3 Diet4 #> #> 1 42 0 0 0 #> 2 51 0 0 0 #> 3 59 0 0 0 #> # ℹ 575 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Constraints on specification for CPA: # 1) The combination of `subject`, `trial`, and `time` must uniquely identify rows in the data # 2) `time` must have constant sampling rate (i.e., evenly spaced values) spec_wrong <- make_jlmer_spec( weight ~ 1 + Diet, ChickWeight, time = \"Time\" ) #> ! Grouping column \"Time\" does not uniquely identify rows in the data #> ! Sampling rate for the `time` column \"Time\" is not constant - may affect interpretability of results. unique(ChickWeight$Time) #> [1] 0 2 4 6 8 10 12 14 16 18 20 21 # Corrected specification for the above spec_correct <- make_jlmer_spec( weight ~ 1 + Diet, subset(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec_correct #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ────────────────────────────────────────────────────────────────────────────────"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":null,"dir":"Reference","previous_headings":"","what":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"Permute data respecting grouping structure(s) observations","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"","code":"permute_by_predictor( jlmer_spec, predictors, predictor_type = c(\"guess\", \"between_participant\", \"within_participant\"), n = 1L )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() predictors vector terms model. multiple, must form levels one predictor. predictor_type Whether predictor \"between_participant\" \"within_participant\". Defaults \"guess\". n Number permuted samples data generate. Defaults 1L.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"long dataframe permuted re-samples .id column representing replication IDs.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_by_predictor.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permute data while respecting grouping structure(s) of observations — permute_by_predictor","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } # Example data setup chickweights_df <- ChickWeight chickweights_df <- chickweights_df[chickweights_df$Time <= 20, ] chickweights_df$DietInt <- as.integer(chickweights_df$Diet) head(chickweights_df) #> Grouped Data: weight ~ Time | Chick #> weight Time Chick Diet DietInt #> 1 42 0 1 1 1 #> 2 51 2 1 1 1 #> 3 59 4 1 1 1 #> 4 64 6 1 1 1 #> 5 76 8 1 1 1 #> 6 93 10 1 1 1 # Example 1: Spec object using the continuous `DietInt` predictor chickweights_spec1 <- make_jlmer_spec( formula = weight ~ 1 + DietInt, data = chickweights_df, subject = \"Chick\", time = \"Time\" ) chickweights_spec1 #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + DietInt #> Predictors: #> DietInt: DietInt #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 4 #> weight DietInt Chick Time #> #> 1 42 1 1 0 #> 2 51 1 1 2 #> 3 59 1 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Shuffling `DietInt` values guesses `predictor_type = \"between_participant\"` reset_rng_state() spec1_perm1 <- permute_by_predictor(chickweights_spec1, predictors = \"DietInt\") #> ℹ Guessed `predictor_type` to be \"between_participant\" # This calls the same shuffling algorithm for CPA in Julia, so counter is incremented get_rng_state() #> [1] 38 # Shuffling under shared RNG state reproduces the same permutation of the data reset_rng_state() spec1_perm2 <- permute_by_predictor(chickweights_spec1, predictors = \"DietInt\") #> ℹ Guessed `predictor_type` to be \"between_participant\" identical(spec1_perm1, spec1_perm2) #> [1] TRUE # Example 2: Spec object using the multilevel `Diet` predictor chickweights_spec2 <- make_jlmer_spec( formula = weight ~ 1 + Diet, data = chickweights_df, subject = \"Chick\", time = \"Time\" ) chickweights_spec2 #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Levels of a category are automatically shuffled together reset_rng_state() spec2_perm1 <- permute_by_predictor(chickweights_spec2, predictors = \"Diet2\") #> ℹ Guessed `predictor_type` to be \"between_participant\" #> ℹ Shuffling all levels of the factor together (\"Diet2\", \"Diet3\", and \"Diet4\") reset_rng_state() spec2_perm2 <- permute_by_predictor(chickweights_spec2, predictors = c(\"Diet2\", \"Diet3\", \"Diet4\")) #> ℹ Guessed `predictor_type` to be \"between_participant\" identical(spec2_perm1, spec2_perm2) #> [1] TRUE # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":null,"dir":"Reference","previous_headings":"","what":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"Simulate cluster-mass statistics via bootstrapped permutations","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"","code":"permute_timewise_statistics( jlmer_spec, family = c(\"gaussian\", \"binomial\"), statistic = c(\"t\", \"chisq\"), nsim = 100L, predictors = NULL, ... )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"jlmer_spec Data prepped jlmer make_jlmer_spec() family GLM family. Currently supports \"gaussian\" \"binomial\". statistic Test statistic calculating cluster mass. Can one \"t\" (default) regression model output \"chisq\" likelihood ratio test (takes twice long calculate). nsim Number simulations description predictors (Optional) subset predictors test. Defaults NULL tests predictors. ... Optional arguments passed Julia model fitting. Defaults fast = TRUE (family = \"binomial\") progress = FALSE.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"simulation--time--predictor 3D array cluster statistics, class timewise_statistics.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/permute_timewise_statistics.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Simulate cluster-mass statistics via bootstrapped permutations — permute_timewise_statistics","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Simulation x Time x Predictor array of t-statistics from regression output reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 3) round(null_statistics, 2) #> , , Predictor = Diet2 #> #> Time #> Sim 0 2 4 6 8 10 12 14 16 18 20 #> 1 -0.90 0.41 0.39 1.12 1.58 2.13 2.34 2.58 2.47 2.67 2.61 #> 2 1.40 0.90 0.96 1.44 1.61 1.45 1.23 0.90 0.80 0.78 0.76 #> 3 -1.44 1.23 0.73 0.88 0.93 0.69 0.58 0.61 0.77 1.04 1.43 #> #> , , Predictor = Diet3 #> #> Time #> Sim 0 2 4 6 8 10 12 14 16 18 20 #> 1 0.00 -0.21 0.11 -0.68 -0.66 -0.38 -0.01 -0.02 -0.17 -0.20 -0.12 #> 2 0.70 0.56 -0.10 0.17 0.37 0.37 0.55 0.21 0.46 0.44 0.02 #> 3 0.72 1.90 0.85 0.20 0.19 -0.09 -0.34 -0.24 0.34 0.44 0.83 #> #> , , Predictor = Diet4 #> #> Time #> Sim 0 2 4 6 8 10 12 14 16 18 20 #> 1 0.45 -0.82 -0.33 -0.68 -0.36 -0.35 -0.41 0.19 -0.04 -0.16 -0.16 #> 2 2.09 -0.69 -0.10 0.34 0.56 0.63 0.35 0.05 -0.05 -0.26 -0.43 #> 3 -2.16 -1.38 -1.35 -0.85 -0.02 -0.12 0.01 0.59 0.71 0.67 0.53 #> # Collect as dataframe with `tidy()` permuted_timewise_stats_df <- tidy(null_statistics) permuted_timewise_stats_df #> # A tibble: 99 × 4 #> predictor time statistic sim #> #> 1 Diet2 0 -8.98e- 1 1 #> 2 Diet2 0 1.40e+ 0 2 #> 3 Diet2 0 -1.44e+ 0 3 #> 4 Diet3 0 2.34e-15 1 #> 5 Diet3 0 6.98e- 1 2 #> 6 Diet3 0 7.21e- 1 3 #> 7 Diet4 0 4.49e- 1 1 #> 8 Diet4 0 2.09e+ 0 2 #> 9 Diet4 0 -2.16e+ 0 3 #> 10 Diet2 2 4.12e- 1 1 #> # ℹ 89 more rows # Permutations ran under the same RNG state are identical reset_rng_state() null_statistics2 <- permute_timewise_statistics(spec, nsim = 3) identical(null_statistics, null_statistics2) #> [1] TRUE get_rng_state() #> [1] 111 null_statistics3 <- permute_timewise_statistics(spec, nsim = 3) identical(null_statistics, null_statistics3) #> [1] FALSE # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/reexports.html","id":null,"dir":"Reference","previous_headings":"","what":"Objects exported from other packages — reexports","title":"Objects exported from other packages — reexports","text":"objects imported packages. Follow links see documentation. generics glance, tidy","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit a Julia regression model using lme4 syntax — to_jlmer","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"Fit Julia regression model using lme4 syntax","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"","code":"to_jlmer( formula, data, family = c(\"gaussian\", \"binomial\"), jlmer_spec_opts = list(), ..., progress = FALSE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"formula Model formula R syntax data data frame family GLM family. Currently supports \"gaussian\" \"binomial\". jlmer_spec_opts List options passed make_jlmer_spec() ... Optional arguments passed Julia model fitting. progress TRUE, prints timing iterations.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"jlmer_mod object.","code":""},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/to_jlmer.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit a Julia regression model using lme4 syntax — to_jlmer","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } jlmerclusterperm_setup(verbose = FALSE) # Fitting a regression model with R formula syntax to_jlmer(weight ~ 1 + Diet, ChickWeight) #> #> ──────────────────────────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) Lower 95% Upper 95% #> ──────────────────────────────────────────────────────────────────────── #> (Intercept) 102.645 4.67395 21.96 <1e-99 93.4847 111.806 #> Diet2 19.9712 7.86744 2.54 0.0111 4.55132 35.3911 #> Diet3 40.3045 7.86744 5.12 <1e-06 24.8847 55.7244 #> Diet4 32.6173 7.91046 4.12 <1e-04 17.113 48.1215 #> ──────────────────────────────────────────────────────────────────────── # `lm()` equivalent summary(lm(weight ~ 1 + Diet, ChickWeight))$coef #> Estimate Std. Error t value Pr(>|t|) #> (Intercept) 102.64545 4.673954 21.961161 4.712762e-78 #> Diet2 19.97121 7.867437 2.538465 1.139708e-02 #> Diet3 40.30455 7.867437 5.122958 4.113938e-07 #> Diet4 32.61726 7.910461 4.123307 4.286352e-05 # Fitting a mixed model with {lme4} syntax to_jlmer(weight ~ 1 + Diet + (1 | Chick), ChickWeight) #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 247.2559 15.7244 #> Residual 4527.9661 67.2902 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 101.781 5.78499 17.59 <1e-68 #> Diet2 20.8353 9.79412 2.13 0.0334 #> Diet3 41.1687 9.79412 4.20 <1e-04 #> Diet4 33.284 9.82881 3.39 0.0007 #> ────────────────────────────────────────────────── # Passing MixedModels.jl fit options to the `...` to_jlmer(weight ~ 1 + Diet + (1 | Chick), ChickWeight, REML = TRUE) #> #> Variance components: #> Column Variance Std.Dev. #> Chick (Intercept) 305.7755 17.4864 #> Residual 4526.5647 67.2797 #> ────────────────────────────────────────────────── #> Coef. Std. Error z Pr(>|z|) #> ────────────────────────────────────────────────── #> (Intercept) 101.634 6.04044 16.83 <1e-62 #> Diet2 20.9832 10.2365 2.05 0.0404 #> Diet3 41.3165 10.2365 4.04 <1e-04 #> Diet4 33.4041 10.27 3.25 0.0011 #> ────────────────────────────────────────────────── # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":null,"dir":"Reference","previous_headings":"","what":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"Test probability cluster-mass statistics range threshold values","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"","code":"walk_threshold_steps( empirical_statistics, null_statistics, steps, top_n = Inf, binned = FALSE, add1 = TRUE, progress = TRUE )"},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"empirical_statistics predictor--time matrix empirical timewise statistics. null_statistics simulation--time--predictor 3D array null (permuted) timewise statistics. steps vector threshold values test top_n many clusters return, order size cluster-mass statistic. Defaults Inf return detected clusters. binned Whether data aggregated/collapsed time bins. Defaults FALSE, requires cluster span least two time points. TRUE, allows length-1 clusters exist. add1 Whether add 1 numerator denominator calculating p-value. Use TRUE effectively count observed statistic part permuted null distribution (recommended larger nsim prior publishing results). progress Whether display progress bar","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"data frame predictor clusters-mass statistics threshold.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/reference/walk_threshold_steps.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Test the probability of cluster-mass statistics over a range of threshold values — walk_threshold_steps","text":"","code":"# \\donttest{ # \\dontshow{ options(\"jlmerclusterperm.nthreads\" = 2) jlmerclusterperm_setup(verbose = FALSE) julia_progress(show = FALSE) # } library(dplyr, warn.conflicts = FALSE) # Specification object spec <- make_jlmer_spec( weight ~ 1 + Diet, filter(ChickWeight, Time <= 20), subject = \"Chick\", time = \"Time\" ) spec #> ── jlmer specification ───────────────────────────────────────── ── #> Formula: weight ~ 1 + Diet2 + Diet3 + Diet4 #> Predictors: #> Diet: Diet2, Diet3, Diet4 #> Groupings: #> Subject: Chick #> Trial: #> Time: Time #> Data: #> # A tibble: 533 × 6 #> weight Diet2 Diet3 Diet4 Chick Time #> #> 1 42 0 0 0 1 0 #> 2 51 0 0 0 1 2 #> 3 59 0 0 0 1 4 #> # ℹ 530 more rows #> ──────────────────────────────────────────────────────────────────────────────── # Compute timewise statistics for the observed and permuted data empirical_statistics <- compute_timewise_statistics(spec) reset_rng_state() null_statistics <- permute_timewise_statistics(spec, nsim = 100) # Test cluster mass/probability under different threshold values walk_threshold_steps(empirical_statistics, null_statistics, steps = 1:3, progress = FALSE) #> # A tibble: 9 × 8 #> threshold predictor id start end length sum_statistic pvalue #> #> 1 1 Diet2 1 2 20 10 19.4 0.00990 #> 2 1 Diet3 1 2 20 10 34.6 0.00990 #> 3 1 Diet4 1 2 20 10 39.4 0.00990 #> 4 2 Diet2 1 4 8 3 8.50 0.0297 #> 5 2 Diet3 1 2 20 10 34.6 0.00990 #> 6 2 Diet4 1 2 20 10 39.4 0.00990 #> 7 3 Diet3 1 4 8 3 12.7 0.00990 #> 8 3 Diet3 2 16 20 3 10.4 0.00990 #> 9 3 Diet4 1 2 12 6 29.7 0.00990 # \\dontshow{ JuliaConnectoR::stopJulia() # } # }"},{"path":[]},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-104","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.4","title":"jlmerclusterperm 1.0.4","text":"CRAN release: 2023-09-06 user-visible changes Restructured internal Julia code module (JlmerClusterPerm.jl) Upon activating project, Manifest.toml cached tools::R_user_dir() speed pre-compilation.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-103","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.3","title":"jlmerclusterperm 1.0.3","text":"CRAN release: 2023-07-19 Fixes CRAN: Use string package version comparison","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-102","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.2","title":"jlmerclusterperm 1.0.2","text":"CRAN release: 2023-07-15","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"bug-fixes-1-0-2","dir":"Changelog","previous_headings":"","what":"Bug fixes","title":"jlmerclusterperm 1.0.2","text":"jlmerclusterperm_setup() now works rc alpha builds Julia Fixes CRAN: exclusion Manifest.toml now specified .Rbuildignore - Project.toml bundled intended.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-101","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.1","title":"jlmerclusterperm 1.0.1","text":"CRAN release: 2023-07-02 jlmerclusterperm_setup() now exits early Julia version requirement (>=1.8) met. Fixes CRAN: Check Julia version requirement proceeding examples tests","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-100","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 1.0.0","title":"jlmerclusterperm 1.0.0","text":"CRAN release: 2023-06-27","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"breaking-changes-1-0-0","dir":"Changelog","previous_headings":"","what":"Breaking changes","title":"jlmerclusterperm 1.0.0","text":"threshold_steps argument walk_threshold_steps() renamed steps.","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"new-features-1-0-0","dir":"Changelog","previous_headings":"","what":"New features","title":"jlmerclusterperm 1.0.0","text":"New functions interface Julia RNG seed: get_rng_seed() set_rng_seed()","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"other-improvements-1-0-0","dir":"Changelog","previous_headings":"","what":"Other improvements","title":"jlmerclusterperm 1.0.0","text":"jlmerclusterperm_setup() now echos Pkg.instantiate() print precompilation information upon first setup call","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-020","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 0.2.0","title":"jlmerclusterperm 0.2.0","text":"Added vignettes. Significant usability improvements","code":""},{"path":"https://yjunechoe.github.io/jlmerclusterperm/news/index.html","id":"jlmerclusterperm-010","dir":"Changelog","previous_headings":"","what":"jlmerclusterperm 0.1.0","title":"jlmerclusterperm 0.1.0","text":"Initial release","code":""}]