Hi Ashley, I think that these two types of analyses are actually asking different things and therefore the decision of which type of analysis is best should be made on biological rather than statistical grounds (i.e., no point in comparing likelihoods or AIC scores). There is a good (albeit, rather technical) discussion of the differences between pGLS and pRMA in the last sections of this paper http://sysbio.oxfordjournals.org/content/61/3/413.short
Sorry this isn't much of answer but hopefully it is somewhat helpful Cheers, Matt On Fri, Oct 3, 2014 at 1:19 PM, Morhardt, Ashley <am159...@ohio.edu> wrote: > Hello everyone, > > I'm working on analyzing data with both pGLS and pRMA in phytools. > Understandably, the two analyses are providing me with slightly different > regression equations, which I need for a subsequent analysis. I am aware > that the output of the pGLS function includes an AIC score. However, for > the pRMA, I see no equivalent test statistic (log-likelihood, AIC, etc) for > the overall model. Is there a way to obtain an equivalent test statistic > for pRMA that would allow me to directly compare these two models? > > In the meantime, I'll be plotting and comparing residual plots to see what > I get. > > Thanks for the help, > > Ashley Morhardt > PhD candidate > Ohio University > _______________________________________________ > 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/ > [[alternative HTML version deleted]] _______________________________________________ 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/
