This generic function calculates the Akaike information criterion for
one or several fitted model objects for which a log-likelihood value
can be obtained, according to the formula $-2 \mbox{log-likelihood}
+ 2 n_{par}$, where $n_{par}$
represents the number of parameters in the fitted model. When comparing
fitted objects, the smaller the AIC, the better the fit.
Usage
## S3 method for class 'lme':
AIC(object, ..., k)
Arguments
object
a fitted model object, for which there exists a
logLik method to extract the corresponding log-likelihood, or
an object inheriting from class logLik.
...
optional fitted model objects.
k
numeric, the ``penalty'' per parameter to be used; the default
k = 2 is the classical AIC.
Value
if just one object is provided, returns a numeric value
with the corresponding AIC; 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 the AIC.
References
Sakamoto, Y., Ishiguro, M., and Kitagawa G. (1986) "Akaike
Information Criterion Statistics", D. Reidel Publishing Company.
data(Orthodont)
fm1 <- lm(distance ~ age, data = Orthodont) # no random effectsAIC(fm1)fm2 <- lme(distance ~ age, data = Orthodont) # random is ~ageAIC(fm1, fm2)