R: Deviance Information Criterion from the threshold model

threshDIC {phytools}

R Documentation

Deviance Information Criterion from the threshold model

This function computes the Deviance Information Criterion from the MCMC object returned by ancThresh.

threshDIC(tree, x, mcmc, burnin=NULL, sequence=NULL, method="pD")

`tree`	phylogenetic tree.
`x`	a named vector containing discrete character states; or a matrix containing the tip species, in rows, and probabilities of being in each state, in columns.
`mcmc`	list object returned by `ancThresh`.
`burnin`	number of generations (not samples) to exclude as burn in; if not supplied then 20 percent of generations are excluded.
`sequence`	assumed ordering of the discrete character state. If not supplied and `x` is a vector then numerical-alphabetical order is assumed; if not supplied and `x` is a matrix, then the column order of `x` is used.
`method`	method for computing the effective number of parameters (options are `"pD"` and `"pV"`).

A vector containing the mean deviance and deviance for the parameter means, the effective number of parameters, and the DIC.

Revell, L. J. (2012) phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol., 3, 217-223.

Revell, L. J. (2014) Ancestral character estimation under the threshold model from quantitative genetics. Evolution, bold68, 743-759.