You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@commons.apache.org by "Christophe S (JIRA)" <ji...@apache.org> on 2015/11/18 15:11:11 UTC

[jira] [Commented] (MATH-1236) Improve Kolmogorov-Smirnov statistic for two samples

    [ https://issues.apache.org/jira/browse/MATH-1236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15011043#comment-15011043 ] 

Christophe S commented on MATH-1236:
------------------------------------

Hi,
Moreover it seems the patch fixes also the problem when we compare two identical distributions (with exactly the same arrays as input). Now the distance is 0, before it wasn't the case!

Thx!

> Improve Kolmogorov-Smirnov statistic for two samples
> ----------------------------------------------------
>
>                 Key: MATH-1236
>                 URL: https://issues.apache.org/jira/browse/MATH-1236
>             Project: Commons Math
>          Issue Type: Improvement
>            Reporter: Otmar Ertl
>             Fix For: 4.0, 3.6
>
>         Attachments: patch
>
>
> I reimplemented the Kolmogorov-Smirnov test statistic for two samples because of its problems with identical values. Although its already fixed in the meantime (see MATH-1197), the proposed code is an improvement, because it is simpler and does not require binary searches.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)