Rev 8184 | Blame | Compare with Previous | Last modification | View Log | Download | RSS feed
% File nlme/man/recalc.corStruct.Rd% Part of the nlme package for R% Distributed under GPL 2 or later: see nlme/LICENCE.note\name{recalc.corStruct}\title{Recalculate for corStruct Object}\usage{\method{recalc}{corStruct}(object, conLin, \dots)}\alias{recalc.corStruct}\alias{recalc.corAR1}\alias{recalc.corARMA}\alias{recalc.corCAR1}\alias{recalc.corCompSymm}%\alias{recalc.corHF} % not implemented\alias{recalc.corNatural}\alias{recalc.corSpatial}\alias{recalc.corSymm}\arguments{\item{object}{an object inheriting from class \code{"\link{corStruct}"},representing a correlation structure.}\item{conLin}{a condensed linear model object, consisting of a listwith components \code{"Xy"}, corresponding to a regression matrix(\code{X}) combined with a response vector (\code{y}), and\code{"logLik"}, corresponding to the log-likelihood of theunderlying model.}\item{\dots}{some methods for this generic require additionalarguments. None are used in this method.}}\description{This method function pre-multiples the \code{"Xy"} component of\code{conLin} by the transpose square-root factor(s) of thecorrelation matrix (matrices) associated with \code{object} and addsthe log-likelihood contribution of \code{object}, given by\code{logLik(object)}, to the \code{"logLik"} component of\code{conLin}.}\value{the recalculated condensed linear model object.}\author{José Pinheiro and Douglas Bates \email{bates@stat.wisc.edu}}\note{This method function is only used inside modelfitting functions, such as \code{lme} and \code{gls},that allow correlated error terms.}\seealso{\code{\link{corFactor}},\code{\link{logLik.corStruct}}}\keyword{models}