Dear Karla,
There are several options for classical phylogenetic regressions. For instance,
the �gls� function from nlme package with the �corStruct� objects from ape, the
function �phylolm� from the package of the same name or �pgls� in caper to name
a few. For multivariate data you can use �mvgls� in mvMORPH.
Note however that the goal of these comparative approaches is to provide valid
confidence interval or p-values associated to parameter estimates rather than
�phylogenetically corrected� slopes or intercepts. Both ordinary least squares
coefficients (conventional non-phylogenetic regressions or OLS) and generalized
least squares (GLS � phylogenetic regressions) might slightly differ but are as
good as any (they�re unbiased - at least assuming that both errors and
predictors are independent and known without errors). However, their variance
may differ substantially and impact the statistical performances of the
regressions.
Best wishes,
Julien
________________________________
De : R-sig-phylo <r-sig-phylo-boun...@r-project.org> de la part de Karla Shikev
<karlashi...@gmail.com>
Envoy� : dimanche 28 juin 2020 15:35
� : R Sig Phylo Listserv <r-sig-phylo@r-project.org>
Objet : [R-sig-phylo] regression slopes
Dear friends,
What would be your suggestion for the best PCM to obtain
"phylogenetically-corrected" slopes and intercepts of a relationship
between two continuous variables?
Karla
[[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/
[[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/
