Dear all, I would like to estimate and to test the significance of the phylogenetic correlation between two binary traits with repeated measures for each species. Actually, one of the traits is the color conspicuousness of a patch and the repetition corresponds actually to the conspicuousness of the different body parts of a bird, so this is different from repeated measured as thought in an measurement error framework. I am mentioning this even if I am not sure how it is important. I am using MCMCglmm but despite many attempts with different priors, the model cannot converge and I end up with improper parameter estimates. In addition, Jarrod Hadfield does not recommend using MCMCglmm with bivariate binary mixed models (see here: https://stat.ethz.ch/pipermail/r-sig-mixed-models/2009q3/002637.html). Is there an alternative to MCMCglmm for doing this in R? Does the phyl.pairedttest from package phytool perform well with binary traits? Thanks in advance for any insight. Best Julien [[alternative HTML version deleted]]
_______________________________________________ R-sig-phylo mailing list R-sig-phylo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
