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
So far following problems with Student-T distribution were identified:
quantile lose precision for p≈0.5 since x becomes close to 1, while p is near zero
quantile::StudentT->Double->Double
quantile (StudentT ndf) p
| p >=0&& p <=1=let x = invIncompleteBeta (0.5* ndf) 0.5 (2*min p (1- p))
incasesqrt$ ndf * (1- x) / x of
r | p <0.5->-r
|otherwise-> r
Both precision and performance suffer greatly for large degrees of freedom. Reason is likely incomplete beta and its inverse perform poorly for large parameters. However distribution becomes close to normal so other approximations could work!
The text was updated successfully, but these errors were encountered:
Quick experimentation with mpmath shows that precision becomes worse for larger NDF. (N of ulps of error proportional to NDF). For very large N mpmath fails itself which I think doesn't bring any good news for algorithm being used...
So far following problems with Student-T distribution were identified:
quantile
lose precision for p≈0.5 since x becomes close to 1, while p is near zeroThe text was updated successfully, but these errors were encountered: