Rev 7351 | Blame | Compare with Previous | Last modification | View Log | Download | RSS feed
\name{gam.fit5.post.proc}\alias{gam.fit5.post.proc}\title{Post-processing output of gam.fit5}\usage{gam.fit5.post.proc(object, Sl, L, lsp0, S, off, gamma)}\arguments{\item{object}{output of \code{gam.fit5}.}\item{Sl}{penalty object, output of \code{Sl.setup}.}\item{L}{matrix mapping the working smoothing parameters.}\item{lsp0}{log smoothing parameters.}\item{S}{penalty matrix.}\item{off}{vector of offsets.}\item{gamma}{parameter for increasing model smoothness in fitting.}}\value{A list containing: \itemize{\item{\code{R}: unpivoted Choleski of estimated expected hessian of log-likelihood. }\item{\code{Vb}: the Bayesian covariance matrix of the model parameters. }\item{\code{Ve}: "frequentist" alternative to \code{Vb}. }\item{\code{Vc}: corrected covariance matrix. }\item{\code{F}: matrix of effective degrees of freedom (EDF). }\item{\code{edf}: \code{diag(F)}. }\item{\code{edf2}: \code{diag(2F-FF)}. }}}\description{INTERNAL function for post-processing the output of \code{gam.fit5}.}\author{Simon N. Wood <simon.wood@r-project.org>.}