You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
E.g. for a set of regression models $s = 1, ..., S$, we can only test that $\beta_{ks} = 0$ for a parameter k. In the example above, the parameter to be tested, "X1", needs to be fixed across models. It is not possible to e.g. test more complex hypotheses, as e.g. $\beta_{1,1} = 0$, $\beta_{1,2} = 0$, etc.
To Do
Define an API that allows users to test more complex hypotheses via bonferroni() and rwolf(). We can draw some inspiration from the rwolf2 Stata module.
Then, implement it =)
The text was updated successfully, but these errors were encountered:
Context
Currently, it is only possible to control for multiple testing for hypothesis of the same parameter, against the same null.
E.g. for a set of regression models$s = 1, ..., S$ , we can only test that $\beta_{ks} = 0$ for a parameter k. In the example above, the parameter to be tested, "X1", needs to be fixed across models. It is not possible to e.g. test more complex hypotheses, as e.g. $\beta_{1,1} = 0$ , $\beta_{1,2} = 0$ , etc.
To Do
bonferroni()
andrwolf()
. We can draw some inspiration from the rwolf2 Stata module.The text was updated successfully, but these errors were encountered: