Hi Pascal, See ?anova.gls.
HTH Emmanuel Le 08/11/2014 17:49, Pascal Title a écrit :
Hi all, I am running a number of PGLS regressions, some of which are multiple regressions. I am using the nlme package with the corBrownian error structure. If I build a model M via multiple regression, then I can get a summary of the model that includes p values as well as a number of other statistics. But how can I get an overall p-value for this model? In the same vein, as value, std. error, t-value and p-value are all returned for each of the traits separately, what does one report for multiple regression models? I have been unable to find a paper that actually reported this type of information for such a model. Here is a toy example: require(nlme) require(phytools) tree <- pbtree(n=20) dat <- as.data.frame(fastBM(tree, nsim=4)) colnames(dat) <- paste('trait', 1:4, sep='') M <- gls(trait1 ~ trait2 + trait3 + trait4, data=dat, correlation=corBrownian(1, tree)) summary(M) Any help/advice would be greatly appreciated. Thanks! -Pascal
_______________________________________________ R-sig-phylo mailing list - R-sig-phylo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/