AICc {gamlr}

Corrected AIC

Description

Corrected AIC calculation.

Usage

AICc(object, k=2)

Arguments

object

Some model object that you can call logLik on (such as a gamlr or glm fit).

k

The usual AIC complexity penalty. k defaults to 2.

Details

This works just like usual AIC, but instead calculates the small sample (or high dimensional) corrected version from Hurvich and Tsai

AICc = -2\log LHD + k*df*\frac{n}{n-df-1}.

Value

A numeric value for every model evaluated.

Author(s)

Matt Taddy [e-mail]

References

Hurvich, C. M. and C-L Tsai, 1989. "Regression and Time Series Model Selection in Small Samples", Biometrika 76.

See Also

gamlr, hockey