This function performs phylogenetic imputation using Maximum Likelihood.
phylo.impute(tree, X, ...)
An object of class "phylo.impute"
consisting of a complete data frame with missing values imputed.
Since optimization is performed numerically using likelihood, a summary of the optimization can be seen by evaluating attr(object,"optim")
, in which object
is of class "phylo.impute"
.
an object of class "phylo"
.
data matrix with species names as row labels. Missing data to be imputed should be coded NA
.
optional arguments.
Liam Revell liam.revell@umb.edu
This function performs phylogenetic imputation in which the evolution of the characters in X
is assumed to have occured by correlation multivariate Brownian motion. Missing values are imputed by maximizing their likelihood jointly with the parameters of the Brownian model. The function evol.vcv
is used internally to compute the likelihood. Note that the Rphylopars package (https://CRAN.R-project.org/package=Rphylopars) also does phylogenetic imputation for multivariate trait data and it seems to be much faster.
Revell, L. J. (2012) phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol., 3, 217-223.
evol.vcv