Re: [Rd] p.adjust(<NA>s), was "Re: [BioC] limma and p-values"

From: Peter Dalgaard <p.dalgaard_at_biostat.ku.dk>
Date: Tue 18 Jan 2005 - 23:57:56 EST

Martin Maechler <maechler@stat.math.ethz.ch> writes:

> MM> I agree that I don't see a good reason to allow specifying 'n'
> MM> as argument unless e.g. for "bonferroni".
> MM> What do other think ?
>
> no reaction yet.
>
> I've thought a bit more in the mean time:
> Assume someone has 100000 P values and knows that he
> only want to adjust the smallest ones.
> Then, only passing the ones to adjust and setting 'n = 100000'
> can be useful and will certainly work for "bonferroni" but
> I think it can't work in general for any other method.
>
> In sum, I still tend to agree that the argument 'n' should be
> dropped -- but maybe with "deprecation" -- i.e. still allow it
> for 2.1.x giving a deprecation warning.

There's another case to consider, namely when you get the same tests multiple times. Think SAS, for instance; when it compares groups you get every comparison twice: I vs III as well as III vs I, so you need a way to say that there are really only k * (k - 1) / 2 tests. Then again, this probably only works for "bonferroni", and R's pairwise.t.test() evades this by extracting the lower.tri before adjustment.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907

______________________________________________
R-devel@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Received on Tue Jan 18 23:07:13 2005

This archive was generated by hypermail 2.1.8 : Tue 18 Jan 2005 - 23:21:26 EST