Hi David. As presently implemented in phylosig(...,method="lambda",test=TRUE), the function tests the null hypothesis that lambda=0. That means that a P-value near zero means that the estimated lambda is highly significantly different from zero. I have not yet implemented the test of lambda=1.0 (which could be interpreted as a test for departure from BM). This can be done by using fitContinuous(...,model="lambda") and fitContinuous(...,model="BM"), and then comparing the likelihoods.
I hope this is helpful. Liam Liam J. Revell University of Massachusetts Boston web: http://faculty.umb.edu/liam.revell/ email: liam.rev...@umb.edu blog: http://phytools.blogspot.com -----Original Message----- From: Martinez Gordillo, David Arturo Sent: Saturday, November 12, 2011 5:13 PM To: r-sig-phylo@r-project.org Subject: [R-sig-phylo] on phylosig Hi Liam I'm working already with phytools really interesting. When using phylosig function with lambda always I get a P=0 like in you example, that means no different to a brownian model? best David _______________________________________________ R-sig-phylo mailing list R-sig-phylo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo _______________________________________________ R-sig-phylo mailing list R-sig-phylo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
