[
https://issues.apache.org/jira/browse/MATH-462?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12975137#action_12975137
]
Phil Steitz commented on MATH-462:
----------------------------------
+1 for including this.
It may take me a little while to fully review the implementation. I am a
little backed at the moment up due to 2.1 release and other stuff. Comments /
review by other [math] community members welcome.
Two things based on quick review:
1) This is s nit, but will make applying the patches a little easier: run
checkstyle (mvn site will do that) and fix the errors that it flags. Missing
braces after if, for example and looks like there are tabs in the source.
These are small issues. By and large, the code fits nicely into Commons Math.
THANKS!
2) Why are the the error tolerances in the tests against the normal and Cauchy
distributions so high? I would expect that with 10000 draws you would get
closer. Unless I misunderstand the parameter setup here, this should work. I
was even going to suggest getting more aggressive and testing some more
quantiles of the empirical distribution of generated values against normal /
Cauchy.
> Random Generator from Stable Distribution
> -----------------------------------------
>
> Key: MATH-462
> URL: https://issues.apache.org/jira/browse/MATH-462
> Project: Commons Math
> Issue Type: New Feature
> Reporter: Pavel Ryzhov
> Priority: Minor
> Fix For: 3.0
>
> Attachments: stable_rnd.patch
>
>
> Stable random generator based on Chambers-Mallows-Stuck method as it is
> described in "Handbook of computational statistics: concepts and methods" by
> James E. Gentle, Wolfgang Härdle, Yuichi Mori
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
