IC: A simple function to extract information criteria from a...

In BPJandree/AutoGLM: Automated fitting, model selection, and documentation functions for GLM modelling of binary response variables.

Description

This function returns a Kullback-Leibler Information Criterion.

Usage

IC(model, KLIC = "AICc", sigma.is.estimated = FALSE)

Arguments

model	A model object that is supported by AIC(). For example produced by lm(), glm(), or speedglm().
KLIC	either "AIC" for the AIC, or "AICc" for the corrected criterion. defaults to "AICc".
sigma.is.estimated	boolean indicating whether variance has been estimated as part of the parameters (Likelihood), or not (LS), If FALSE, number of parameters is increased by 1 for comparison with MLE objects. Defaults to FALSE. Should not impact the results when models of the same type are compared, but the option has been added to allow users to change the number of counted parameters when comparing MLE results with LS results.

Value

numeric value.

Examples

somepars <- c(-0.5, 0.4, -0.3, 0.2, -0.1, 0, 0, 0)
someData <- simulateLogit(nobs=2500, pars=somepars)
someModel <- logit(someData)
IC(someModel, KLIC="AIC")
AIC(someModel) # compare
IC(someModel) # AICc

BPJandree/AutoGLM documentation built on May 5, 2019, 10:25 a.m.