[R-sig-phylo] R-squared alternative for gls
Hello members, I'm creating relative values os semicircular canal size by fitting a PGLS and extracting residuals (using phyl.resid). I've heard that R-squared isn't meaningful in gls models. What I'm trying to do is to know is which of the two independent variables is best and on a lm () model I would just check the R-squared. Is the alternative, in the case of gls(), looking at the AIC or logLik? Thanks in advance. Best regards, Sérgio. -- Com os melhores cumprimentos, Sérgio Ferreira Cardoso. Best regards, Sérgio Ferreira Cardoso MSc. Paleontology candidate Departamento de Ciências da Terra - FCT /Universidade Nova de Lisboa Geociências - Universidade de Évora Lisboa, Portugal ᐧ [[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/
Re: [R-sig-phylo] Different Lambda estimations
Hi Emmanuel, Thanks for your answer. So that means I can still use that Lambda estimated value, I suppose. In the case of Lambda1, it then means that species are more similar than expected, right? Best regards, Sérgio. 2015-07-17 15:57 GMT+01:00 Emmanuel Paradis emmanuel.para...@ird.fr: Hi Sérgio, In phytools, lambda is constrained to be positive, not in ape. A negative value of lambda is the consequence of closely related species being more dissimilar than expected under phylogenetic dependence (lambda = 1) or under independence (lambda = 0). This can be interpreted biologically as close species diverging into distinct niches and/or phenotypes. Best, Emmanuel Le 16/07/2015 20:34, Sergio Ferreira Cardoso a écrit : Hello all, I'm trying to estimate a lambda for my phylogeny but something strange happened. I used 3 different packages to confirm the estimated lambda value. My tree is ultrametric (with branchlengths in million years), it has 59 tips and 58 internal nodes. Here is what I did: df-data.frame(y,x) df$Species-rownames(df) tree-read.nexus(TTOL_tree.nex) vf-diag(vcv(tree)) phyl.resid(tree,x,y,method=lambda) ## phytools lambda = 0.741795 fit-gls(y~x, correlation = corPagel(value=1,phy=tree,fixed=FALSE), data = df, method=ML,weights=varFixed(~vf)) ## ape nlme lambda = -0.6863758 fit-pgls(y~x,comparative.data(tree,df,Species), lambda=ML) ## caper lambda = 0.742 Does anyone know why ape's gls estimates a value as different as that? Lambda varies between 0 and 1. Is there any other way to estimate Lambda (just to make another check). This is not the first time Lambda estimations of my trees deliver values as strange as this one... Cheers, Sérgio. -- Com os melhores cumprimentos, Sérgio Ferreira Cardoso. Best regards, Sérgio Ferreira Cardoso MSc. Paleontology candidate Departamento de Ciências da Terra - FCT /Universidade Nova de Lisboa Geociências - Universidade de Évora Lisboa, Portugal [[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/
Re: [R-sig-phylo] R-squared alternative for gls
It's not that r squared isn't meaningful for generalized least squares but rather that it cannot be compared directly with values from OLS models. Cheers, Ted From: R-sig-phylo [r-sig-phylo-boun...@r-project.org] on behalf of Sergio Ferreira Cardoso [sff.card...@campus.fct.unl.pt] Sent: Monday, July 20, 2015 4:11 AM To: R phylo mailing list mailing list Subject: [R-sig-phylo] R-squared alternative for gls Hello members, I'm creating relative values os semicircular canal size by fitting a PGLS and extracting residuals (using phyl.resid). I've heard that R-squared isn't meaningful in gls models. What I'm trying to do is to know is which of the two independent variables is best and on a lm () model I would just check the R-squared. Is the alternative, in the case of gls(), looking at the AIC or logLik? Thanks in advance. Best regards, Sérgio. -- Com os melhores cumprimentos, Sérgio Ferreira Cardoso. Best regards, Sérgio Ferreira Cardoso MSc. Paleontology candidate Departamento de Ciências da Terra - FCT /Universidade Nova de Lisboa Geociências - Universidade de Évora Lisboa, Portugal ᐧ [[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/ ___ 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/
Re: [R-sig-phylo] R-squared alternative for gls
Dear Dr. Garland, Then there is no problem because I just need to compare between phylogenetic models. Thank you. Best regards, Sérgio. 2015-07-20 14:48 GMT+01:00 Theodore Garland Jr theodore.garl...@ucr.edu: It's not that r squared isn't meaningful for generalized least squares but rather that it cannot be compared directly with values from OLS models. Cheers, Ted From: R-sig-phylo [r-sig-phylo-boun...@r-project.org] on behalf of Sergio Ferreira Cardoso [sff.card...@campus.fct.unl.pt] Sent: Monday, July 20, 2015 4:11 AM To: R phylo mailing list mailing list Subject: [R-sig-phylo] R-squared alternative for gls Hello members, I'm creating relative values os semicircular canal size by fitting a PGLS and extracting residuals (using phyl.resid). I've heard that R-squared isn't meaningful in gls models. What I'm trying to do is to know is which of the two independent variables is best and on a lm () model I would just check the R-squared. Is the alternative, in the case of gls(), looking at the AIC or logLik? Thanks in advance. Best regards, Sérgio. -- Com os melhores cumprimentos, Sérgio Ferreira Cardoso. Best regards, Sérgio Ferreira Cardoso MSc. Paleontology candidate Departamento de Ciências da Terra - FCT /Universidade Nova de Lisboa Geociências - Universidade de Évora Lisboa, Portugal ᐧ [[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/ -- Com os melhores cumprimentos, Sérgio Ferreira Cardoso. Best regards, Sérgio Ferreira Cardoso MSc. Paleontology candidate Departamento de Ciências da Terra - FCT /Universidade Nova de Lisboa Geociências - Universidade de Évora Lisboa, Portugal [[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/