I have a quick question: I'm running a PGLS with many correlated traits and
my co-author suggested to first apply PCA on the traits, then use the axes
as predictors in my PGLS.

I was wondering whether I need to perform a phylo PCA or if a regular PCA
would do the trick? My intuition is that pPCA would be better as it takes
phylogeny into account, but again it isn't being estimated jointly with the
rest of the PGLS model...

Also, is there any theory as to how this will impact phylogenetic signal in
the residuals of the model? For some reason it feels odd to take phylogeny
into account twice...


        [[alternative HTML version deleted]]

R-sig-phylo mailing list - R-sig-phylo@r-project.org
Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/

Reply via email to