[jira] [Commented] (MATH-1233) Uncommon wilcoxon signed-rank p-values

["Phil Steitz (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MATH-1233?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16458905#comment-16458905 ] 

Phil Steitz commented on MATH-1233:
-----------------------------------

A first attempt at a fix for this was implemented in [this issue|https://github.com/Hipparchus-Math/hipparchus/issues/37] in Hipparchus.  Current code there discards tied pairs, which would result in an exception if all pairs are tied.  I agree that it should be configurable how ties are handled and the scipy alternatives make sense.  Hipparchus currently implements the simplest one, what scipy calls "wilcox."  Another problem with the current [math] implementation is that the continuity correction is not applied correctly.  That was fixed in the Hipparchus patch, which should backport easily.

> Uncommon wilcoxon signed-rank p-values
> --------------------------------------
>
>                 Key: MATH-1233
>                 URL: https://issues.apache.org/jira/browse/MATH-1233
>             Project: Commons Math
>          Issue Type: Bug
>    Affects Versions: 3.5
>            Reporter: Icaro Cavalcante Dourado
>            Assignee: Rob Tompkins
>            Priority: Major
>             Fix For: 4.0
>
>         Attachments: MATH-1233-test.patch
>
>
> This implementation in WilcoxonSignedRankTest looks weird. For equal vectors, the correct pValue should be 1, because it is the probability of the vectors to come from same population.
> On the opposite, this implementation returns ~0 for equal vectors. So we need to analyze the returned pValue > significanceLevel to reject H0 hypothesis, while in R and many others tools we perform the opposite: pValue <= significanceLevel gives us an argument to reject null hypothesis.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)