Displaying 2 results from an estimated 2 matches for "aakinshin".
2024 Feb 12
0
Errors in wilcox family functions
...ecreases the error in p-value approximation without substantially rewriting the internals. An example patch and benchmarks are available at https://github.com/ahl27/R_Patches/tree/94e8e0bcf5076841637f1031ea9cf456ad18d7ef/EdgeworthExpansion.
More verbose details, examples, and benchmarks:
- https://aakinshin.net/posts/r-mann-whitney-incorrect-p-value/
- https://aakinshin.net/posts/mw-edgeworth/
2. The built-in Hodges-Lehmann estimator has some edge cases that result in incorrect answers without suitable warnings. These are detailed extensively at the links below. In short, the following cases result i...
2023 Dec 11
1
Base R wilcox.test gives incorrect answers, has been fixed in DescTools, solution can likely be ported to Base R
...answers (see
<https://github.com/AndriSignorell/DescTools/issues/97>
https://github.com/AndriSignorell/DescTools/issues/97). The error is driven
by the existence of tied values forcing wilcox.test in Base R to switch to
an approximate algorithm that returns incorrect results - see
<https://aakinshin.net/posts/r-hodges-lehmann-problems/>
https://aakinshin.net/posts/r-hodges-lehmann-problems/ for a detailed
exposition of the issue.
Andri Signorell and Cyril Moser have a new C++ implementation of
DescTools::HodgesLehmann using a O(N log(N)) algorithm due to Monahan, but
wilcox.test in Base...