AIC.bellreg {bellreg}

Akaike information criterion

Description

Akaike information criterion

Usage

## S3 method for class 'bellreg'
AIC(object, ..., k = 2)

Arguments

object

an object of the class bellreg.

...

further arguments passed to or from other methods.

k

numeric, the penalty per parameter to be used; the default k = 2 is the classical AIC.

Value

the Akaike information criterion value when a single model is passed to the function; otherwise, a data.frame with the Akaike information criterion values and the number of parameters is returned.

Examples

library(bellreg)
data(faults)
fit1 <- bellreg(nf ~ 1, data = faults, approach = "mle")
fit2 <- bellreg(nf ~ lroll, data = faults, approach = "mle")
AIC(fit1, fit2)