Thanks for all the replies,
I tried the OU transform as suggested by Tony and it gives a consistent and
more typical result (independent of starting value), so I guess that is an
option.
Given Liam's first comment, I am wondering whether taking things like negative
lambda as a quirk or not
http://nere.bio.uci.edu/
(A University of California Multicampus Research Project)
Original message
Date: Thu, 6 Aug 2009 23:04:08 -0400 (EDT)
From: Liam J. Revell lrev...@fas.harvard.edu
Subject: Re: [R-sig-phylo] negative lambda in gls with
corPagel?
To: Gi-Mick