https://github.com/cran/grplasso
Tip revision: 4aa9c8aef3fbd34e15d7e1ab26d5f96d7992b62c authored by Lukas Meier on 07 May 2020, 15:20:02 UTC
version 0.4-7
version 0.4-7
Tip revision: 4aa9c8a
grpl.control-class.Rd
\name{grpl.control-class}
\docType{class}
\alias{grpl.control-class}
\title{Class "grpl.control": Options for the Group Lasso Algorithm}
\description{
Objects of class "grpl.control" define options such as bounds on the Hessian,
convergence criteria and output management for the Group Lasso algorithm.
}
\section{Objects from the Class}{
Objects can be created by calls of the form \code{grpl.control(...)}
}
\section{Slots}{
\describe{
\item{\code{save.x}}{a logical indicating whether the design matrix
should be saved.}
\item{\code{save.y}}{a logical indicating whether the response should
be saved.}
\item{\code{update.hess}}{should the hessian be updated in each
iteration ("always")? update.hess = "lambda" will update
the Hessian once for each component of the penalty
parameter "lambda" based on the parameter estimates
corresponding to the previous value of the penalty
parameter.}
\item{\code{update.every}}{Only used if update.hess = "lambda". E.g. set to 3
if you want to update the Hessian only every third grid point.}
\item{\code{inner.loops}}{How many loops should be done (at maximum)
when solving only the active set (without considering the remaining
predictors). Useful if the number of predictors is large. Set to 0
if no inner loops should be performed.}
\item{\code{line.search}}{Should line searches be performed?}
\item{\code{max.iter}}{Maximal number of loops through all groups}
\item{\code{tol}}{convergence tolerance; the smaller the more precise.}
\item{\code{lower}}{lower bound for the diagonal approximation of the
corresponding block submatrix of the Hessian of the negative
log-likelihood function.}
\item{\code{upper}}{upper bound for the diagonal approximation of the
corresponding block submatrix of the Hessian of the negative
log-likelihood function.}
\item{\code{beta}}{scaling factor \eqn{\beta < 1} of the Armijo line search.}
\item{\code{sigma}}{\eqn{0 < \sigma < 1} used in the Armijo line search.}
\item{\code{trace}}{integer. \code{1} prints the current lambda value,
\code{2} prints the improvement in the objective function after each
sweep through all the parameter groups and additional information.}
}
}
\details{For the convergence criteria see chapter 8.2.3.2 of Gill et
al. (1981).}
\references{Philip E. Gill, Walter Murray and Margaret H. Wright (1981)
\emph{Practical Optimization}, Academic Press.
Dimitri P. Bertsekas (2003) \emph{Nonlinear Programming}, Athena Scientific.}
\keyword{classes}