[R] Post hoc tests

[Peter Dalgaard BSA]

myotis at cix.compulink.co.uk (Graham Smith) writes:

> This is only partly an R question, but this seems a good place to ask.
> 
> Like many, if not most, statistics programs, R has no post hoc tests 
> for Kruskal Wallis or Friedman, and as far as I can make out they are 
> not available in any of the R packages. I would be grateful if someone 
> could point me in the right direction if I am wrong.

I think pairwise.wilcox.test is just about as close as one gets for
the KW case. 

> As I said this is not uncommon. Parametric Anova procedures almost 
> always, if not always, come with a post hoc procedure. Why is it not 
> the same with non-parametric anovas.

The math doesn't work out as neatly... E.g. the distribution of rank
sums with some groups different is largely unknown. Er, well, it can
be worked out as a permutation distribution I suppose, but it isn't
nice and to my knowledge there's nothing that corresponds to the
studentized range distribution. Then again, maybe there could be.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._