[R] Two-way non-parametric ANOVA?

[Matthias Gondan]

Hi Diana,

Look at Edgar Brunner's Homepage

http://www.ams.med.uni-goettingen.de/en/sta/e.brunner.html

and here:

http://www.ams.med.uni-goettingen.de/en/sof/index.html

Unfortunately, the R script for a 2x2 design with two groups and two
timepoints (f1.ld.f1.r) seems to be broken. Uncommenting the branch to
the special case "case2x2" in the script should do well, and more than 2
groups or timepoints seem to be working, too.

The basic problem with non-parametric (i.e. ordinal) Anova with more
than one factor is that the covariance matrix is not homoscedastic
anymore after a rank transformation, and autoregressive or compound
symmetry covariance structures are destroyed.

Best wishes,

Matthias

Diana Perdiguero schrieb:
> We need a two-way non-parametric ANOVA in order to analysis properly some
> ecological data, do you know any reference in R? or how to do it?
> Thank you very much
> All the best
>
> diana
>