DIC {OrthoPanels}

Deviance Information Criterion (DIC)

Description

Computes the Deviance Information Criterion (DIC), which is a generalization of the Akaike Information Criterion. Models with smaller DIC are considered to fit better than models with larger DIC.

Usage

DIC(object, ...)

Arguments

object

an instance of class opm whose DIC is wanted.

...

further arguments passed to other methods.

Details

DIC is defined as DIC = 2*\bar{D} - D_\theta where: \bar{D} = -2 mean(log-likelihood at parameter samples) D_\theta = -2 * log(likelihood at expected value of parameters)

DIC is calculated as: 2 * (-2 * mean(log-likelihood at each element of parameter samples)) - (-2 * log(likelihood at mean parameter sample value))

Value

a numeric value with the corresponding DIC

Note

Note the speed of computation of the DIC in proportional to the number of sampled values of the parameters in the opm object.