[R-sig-phylo] randomizing clades within a node

2012-01-13 Thread Juan Antonio Balbuena
Hello I would very much appreciate if someone can help me with this one. I wish to compare two additive trees that are identical for some but not all the clades. So I use rtree of the ape package with, say, 10 tips to generate tree "A". Then I wish to

[R-sig-phylo] Lambda; PIC vs. PGLS

2012-01-13 Thread Karin Schneeberger
Dear all I am currently analysing a set of traits from different species - let's say brain size and body mass - using a tree where branch lengths were set artificially (after Graphen) as the length was not available.  I calculate phylogenetic independent contrasts (PIC) using the package caper

Re: [R-sig-phylo] Lambda; PIC vs. PGLS

2012-01-13 Thread Liam J. Revell
Hi Karin. gls(...,correlation=corPagel(...,fixed=F),method=ML) and pgls(...,lambda=ML) should produce the same result unless one or the other is not optimizing properly or unless different bounds on the fitted model parameters (in particular, lambda) are used. To show this to yourself, you