MuMIn (version 1.3.6)

AICc: Second-order Akaike Information Criterion

Description

Calculates second-order Akaike information criterion for one or several fitted model objects (AIC for small samples).

Usage

AICc(object, ..., k = 2, REML = NULL)

Arguments

object

a fitted model object for which there exists a logLik method, or a logLik object

...

optionally more fitted model objects

k

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

REML

optional logical value, passed to the logLik method indicating whether the restricted log-likelihood or log-likelihood should be used. The default is to use the method used for model estimation.

Value

• If just one object is provided, returns a numeric value with the corresponding AICc; if more than one object are provided, returns a data.frame with rows corresponding to the objects and columns representing the number of parameters in the model (df) and AICc.

encoding

utf-8

References

Burnham, K. P. and Anderson, D. R (2002) Model selection and multimodel inference: a practical information-theoretic approach. 2nd ed.

See Also

Akaike's An Information Criterion: AIC

AICc in package AICcmodavg, aicc in package glmulti

Examples

#Model-averaging mixed models

library(nlme)
data(Orthodont, package = "nlme")

# Fit model by REML
fm2 <- lme(distance ~ Sex*age + age*Sex, data = Orthodont,
    random = ~ 1|Subject / Sex, method = "REML")

# Model selection: ranking by AICc using ML
dd <- dredge(fm2, trace=TRUE, rank="AICc", REML=FALSE)

(attr(dd, "rank.call"))

# Get the models (fitted by REML, as in the global model)
gm <- get.models(dd, 1:4)

# Because the models originate from 'dredge(..., rank=AICc, REML=FALSE)', 
# the default weights in 'model.avg' are ML based:
model.avg(gm, method = "NA")
# same result
#model.avg(gm, method = "NA", rank="AICc", rank.args = list(REML=FALSE))
# REML based weights
model.avg(gm, method = "NA", rank="AICc", rank.args = list(REML=TRUE))