commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Thorsten Schäfer (JIRA) <j...@apache.org>
Subject [jira] [Commented] (MATH-1034) Add binomial test
Date Mon, 30 Sep 2013 22:20:24 GMT

    [ https://issues.apache.org/jira/browse/MATH-1034?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13782341#comment-13782341
] 

Thorsten Schäfer commented on MATH-1034:
----------------------------------------

Thanks Phil!

Regarding improvement 1: I tried this in my first implementation attempt, but it lead to wrong
results. When I read more about how the test value is calculated, I understood it as follows:
your suggested approach would work in a case where p=0.5 as the distribution is then symmetrical.
However, for other values this is not the case. For instance, if we test with p=0.99, there's
basically no room on the right-side for errors, but a lot of room on the left side. The test
tried to capture the most extreme outliners for a given significance level and in this example
they will be most likely all on the left side - so just considering one side and doubling
should not work. (At least that's what my understanding is now...)

> Add binomial test
> -----------------
>
>                 Key: MATH-1034
>                 URL: https://issues.apache.org/jira/browse/MATH-1034
>             Project: Commons Math
>          Issue Type: Wish
>    Affects Versions: 3.2
>            Reporter: Thorsten Schäfer
>            Priority: Minor
>         Attachments: binomialTest.patch, binomialTest.patch
>
>
> A binomial test would be a nice addition to commons-math. I might supply a patch in the
near future. I guess the interface should be similar to the other tests, i.e., a method to
get the p-value and a method returning a boolean indicating reject/non-reject.
> Is there a policy about using Enumerations in commons-math? For instance, in R you can
test two-sided, less or greater. This could be done using an enumeration in Java, but I'm
not sure if this is discouraged for backward compatibility reasons...



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message