[RkWard-devel] wilcoxon & ansari_bradley tests

[SJR]

Am Mittwoch, 7. Februar 2007 12:49 schrieb Thomas Friedrichsmeier:
> On Sunday 04 February 2007 20:24, SJR wrote:
> > > - Does this note recommed turning on the "compute exact p-value"
> > > option, or does it recommed to use the exact version of the test? I
> > > don't have any background knowledge on the difference between these
> > > two.
> >
> > By default an asymptotic p-value is calculated but it's recommended to
> > use the exact option (within this test) if there are less then 50 values
> > and if there is no binding (Don't know if this is the right term German:
> > "Bindung"). If n < 50 and if there is binding one should use the
> > wilcoxon.exact() test (exactRankTests package) for correct results. Hope
> > this answers your question.
>
> I see. One last suggestion then: In the documentation for wilcox.test and
> ansari.test, it says "By default (if exact is not specified), an exact
> p-value is computed if both samples contain less than 50 finite values and
> there are no ties. Otherwise, a normal approximation is used." So why not
> just make use of this. I.e. instead of a checkbox, provide a radio with
> three options:
>
> Compute exact p-value
> * Yes
> * Automatic
> * No
>
> In the automatic case (default), the exact=... parameter would be omitted.
>
> Does this sound reasonable?

More that that. It sounds clever.

>
> Regards
> Thomas