[pymvpa] How to do a simple t or F test in PyMVPA

[Michael Hanke]

Hi,

On Sat, May 30, 2009 at 08:29:35AM +0200, Michael Hanke wrote:
<snip>
> I guess you are right. OneWayAnova() would be the way to generate
> F-scores, but it looks like it doesn't not return properly F-distributed
> scores, because it does not consider the degrees of freedom --
> unfortunately, it is probably me who is responsible :(

<snip>

> Moreover, I will fix OneWayAnova to use SciPy as well. For additional
> tests also scipy.stats is probably the best source.

I have just pushed a new implementation of OneWayAnova() that should
produce proper F-scores. It does not use SciPy, but it is based on its
implementation. I also added a unittest for compliance with SciPy's
f_oneway() function to make sure this doesn't happen again.

I pushed the pacth into my branch and also cherry-picked it into
'master'.


Michael


-- 
GPG key:  1024D/3144BE0F Michael Hanke
http://apsy.gse.uni-magdeburg.de/hanke
ICQ: 48230050