% Generated by roxygen2: do not edit by hand

% Please edit documentation in R/projection.R

\name{projsel}

\alias{projsel}

\title{Forward selection minimizing KL-divergence in projection}

\usage{

projsel(obj, max.iters = 30, start.from = NULL, out.csv = NULL)

}

\arguments{

\item{obj}{Object of class \code{hsstan}.}

\item{max.iters}{Maximum number of iterations (number of predictors selected)

after which the selection procedure should stop.}

\item{start.from}{Vector of variable names to be used in the starting

submodel. If \code{NULL} (default), selection starts from the set of

unpenalized covariates if the model contains penalized predictors,

otherwise selection starts from the intercept-only model.}

\item{out.csv}{If not \code{NULL}, the name of a CSV file to save the

output to.}

}

\value{

A data frame of class \code{projsel} where each row corresponds to a

forward-selected submodel that contains all variables listed up to that row.

Attribute \code{start.from} reports the predictors in the initial model.

The data frame contains the following columns:

\item{var}{names of the variables selected.}

\item{kl}{KL-divergence from the full model to the submodel.}

\item{rel.kl.null}{relative explanatory power of predictors starting from the

intercept-only model.}

\item{rel.kl}{relative explanatory power of predictors starting from the

initial submodel.}

\item{elpd}{the expected log predictive density of the submodels.}

\item{delta.elpd}{the difference in elpd from the full model.}

}

\description{

Forward selection minimizing KL-divergence in projection

}

\examples{

\donttest{

\dontshow{utils::example("hsstan", echo=FALSE)}

\dontshow{oldopts <- options(mc.cores=2)}

# continued from ?hsstan

sel <- projsel(hs.biom, max.iters=3)

plot(sel)

\dontshow{options(oldopts)}

}

}