2
%\docType{genericFunction}
4
\alias{BIC,logLik-method}
6
\title{Bayesian Information Criterion}
8
The \code{\link[nlme]{BIC}} generic function calculates the Bayesian
9
information criterion, also known as Schwarz's Bayesian criterion
10
(SBC), for one or several fitted model objects for which a
11
log-likelihood value can be obtained, according to the formula \eqn{-2
12
\mbox{log-likelihood} + n_{par} \log(n_{obs})}{-2*log-likelihood +
13
npar*log(nobs)}, where \eqn{n_{par}}{npar} represents the number of
14
parameters and \eqn{n_{obs}}{nobs} the number of observations in the
21
% \item{object}{An object of a suitable class for the BIC to be
22
% calculated - usually a \code{\link[stats]{logLik}} object
23
% created by a call to the \code{\link[stats]{logLik}} generic.
25
% \item{\dots}{Some methods for this generic function may take
26
% additional, optional arguments. At present none do.}
29
if just one object is provided, returns a numeric value with the
30
corresponding BIC; if more than one object are provided, returns a
31
\code{data.frame} with rows corresponding to the objects and columns
32
representing the number of parameters in the model (\code{df}) and the
37
Estimating the Dimension of a Model,
38
\emph{Annals of Statistics} \bold{6}, 461--464.
40
\seealso{\code{\link[nlme]{BIC}}, \code{\link[stats]{logLik}},
41
\code{\link[stats]{AIC}}}