mvMORPH (version 1.1.0)

GIC: Generalized Information Criterion (GIC) to compare models fit with mvgls by Maximum Likelihood (ML) or Penalized Likelihood (PL).

Description

The GIC (Konishi & Kitagawa 1996) allows comparing models fit by Maximum Likelihood (ML) or Penalized Likelihood (PL).

Usage

GIC(object, ...)

Arguments

object

An object of class 'mvgls'. See ?mvgls

...

Options to be passed through.

Value

a list with the following components

LogLikelihood

the log-likelihood estimated for the model with estimated parameters

GIC

the GIC criterion

bias

the value of the bias term estimated to compute the GIC

Details

The Generalized Information Criterion (GIC) allows comparing the fit of various models estimated by Penalized Likelihood (see ?mvgls). See also the gic_criterion function in the RPANDA package. Note that the current implementation of the criterion has not been tested for multiple predictors comparison. Prefer simulation based comparisons instead.

References

Clavel, J., Aristide, L., Morlon, H., 2018. A Penalized Likelihood framework for high-dimensional phylogenetic comparative methods and an application to new-world monkeys brain evolution. Systematic Biology DOI:10.1093/sysbio/syy045.

Konishi S., Kitagawa G. 1996. Generalised information criteria in model selection. Biometrika. 83:875-890.

See Also

mvgls mvgls.pca

Examples

# NOT RUN {
set.seed(1)
n <- 32 # number of species
p <- 50 # number of traits

tree <- pbtree(n=n) # phylogenetic tree
R <- crossprod(matrix(rnorm(p*p), ncol=p)) # a random symmetric matrix (covariance)
# simulate a dataset
Y <- mvSIM(tree, model="BM1", nsim=1, param=list(sigma=R))

fit1 <- mvgls(Y~1, tree=tree, model="BM", method="H&L")
fit2 <- mvgls(Y~1, tree=tree, model="OU", method="H&L")


GIC(fit1); GIC(fit2)
# }