commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Christian Winter (Issue Comment Edited) (JIRA)" <j...@apache.org>
Subject [jira] [Issue Comment Edited] (MATH-699) inverseCumulativeDistribution fails with cumulative distribution having a plateau
Date Sun, 04 Dec 2011 20:38:39 GMT

    [ https://issues.apache.org/jira/browse/MATH-699?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13162458#comment-13162458
] 

Christian Winter edited comment on MATH-699 at 12/4/11 8:37 PM:
----------------------------------------------------------------

Yes, we can get rid of {{IntegerDistribution.getDomainLowerBound(double)}} and {{IntegerDistribution.getDomainUpperBound(double)}}.
I will do this in the reimplementation of {{IntegerDistribution.inverseCumulativeProbability(double)}}
based on MATH-692.

Using Chebyshev's inequality also can be adopted. But a plateau detection step is not needed
there because {{IntegerDistribution}} requires its own solver for finding the inverse cumulative
probability anyway. Thus this solver can apply the correct inequalities for finding the "right"
quantile automatically. (Using the "correct" inequalies for the solver in {{RealDistribution}}
wasn't an option because an existing general purpose solver is utilized there.)
                
      was (Author: cwinter):
    Yes, we can get rid of {{IntegerDistribution.getDomainLowerBound(double)}} and {{IntegerDistribution.getDomainUpperBound(double)}}.
I will do this in the reimplementation of {{IntegerDistribution.inverseCumulativeProbability(double)}}
based on MATH-692.
                  
> inverseCumulativeDistribution fails with cumulative distribution having a plateau
> ---------------------------------------------------------------------------------
>
>                 Key: MATH-699
>                 URL: https://issues.apache.org/jira/browse/MATH-699
>             Project: Commons Math
>          Issue Type: Bug
>    Affects Versions: 2.2
>            Reporter: S├ębastien Brisard
>            Assignee: S├ębastien Brisard
>            Priority: Minor
>             Fix For: 3.0
>
>         Attachments: AbstractContinuousDistributionTest.java, inv-cum-new-impl.zip
>
>
> This bug report follows MATH-692. The attached unit test fails. As required by the definition
in MATH-692, the lower-bound of the interval on which the cdf is constant should be returned.
This is not so at the moment.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

       

Mime
View raw message