[jira] [Updated] (MATH-1245) Kolmogorov-Smirnov exactP gives incorrect p-values for some D-statistics

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

     [ https://issues.apache.org/jira/browse/MATH-1245?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Phil Steitz updated MATH-1245:
------------------------------
    Affects Version/s: 3.5

> Kolmogorov-Smirnov exactP gives incorrect p-values for some D-statistics
> ------------------------------------------------------------------------
>
>                 Key: MATH-1245
>                 URL: https://issues.apache.org/jira/browse/MATH-1245
>             Project: Commons Math
>          Issue Type: Bug
>    Affects Versions: 3.4.1, 3.5
>            Reporter: Phil Steitz
>            Priority: Minor
>             Fix For: 4.0, 3.6
>
>
> The exactP method in KolmogorovSmirnovTest, which is used by default for small samples in 2-sample tests, can give slightly incorrect p-values in some cases.  The reason for this is that p(D > d) is computed by examining all m-n partitions and counting the number of partitions that give D values larger than the observed value.  D values are not rounded, so some values that are mathematically identical to the observed value compare less than or greater than it.  This results in small errors in the reported p-values.



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