https://github.com/cran/MuMIn
Tip revision: b4fae7a102a8a67407ddf3b3ef825ed34735704b authored by Kamil Bartoń on 18 December 2019, 19:20:02 UTC
version 1.43.14
version 1.43.14
Tip revision: b4fae7a
get.models.Rd
\name{get.models}
\alias{get.models}
\alias{pget.models}
\encoding{utf-8}
\title{Retrieve models from selection table}
\description{
Generate or extract a list of fitted model objects from a
\code{"model.selection"} table, optionally using parallel computation in a
cluster.
}
\usage{
get.models(object, subset, cluster = NA, ...)
}
\arguments{
\item{object}{ object returned by \code{\link{dredge}}. }
\item{subset}{subset of models, an expression evaluated within the model
selection table (see \sQuote{Details}).}
\item{cluster}{ optionally, a \code{"cluster"} object. If it is a valid
cluster, models are evaluated using parallel computation. }
\item{\dots}{additional arguments to update the models. For example, in
\code{lme} one may want to use \code{method = "REML"} while using \code{"ML"}
for model selection.}
}
\value{
\code{\link{list}} of fitted model objects.
}
\details{
The argument \code{subset} must be explicitely provided. This is to assure that
a potentially long list of models is not fitted unintentionally. To evaluate all
models, set \code{subset} to \code{NA} or \code{TRUE}.
If \code{subset} is a character vector, it is interpreted as names of rows to be
selected.
}
\note{
Alternatively, \code{getCall} and \code{eval} can be used to compute a model out of the
\code{"model.selection"} table (e.g. \code{eval(getCall(<model.selection>, i))}, where
\code{i} is the model index or name).
Using \code{get.models} following \code{dredge} is not efficient as the requested models
have to be fitted again. If the number of generated models is reasonable, consider using \cr
\code{lapply(dredge(..., evaluate = FALSE), eval)}, which generates a list of all model
calls and evaluates them into a list of model objects. This avoids fitting the
models twice.
\code{pget.models} is still available, but is deprecated.
}
\author{Kamil Barto\enc{ń}{n}}
\seealso{
\code{\link{dredge}} and \code{\link{pdredge}}, \code{\link{model.avg}}
\code{\link{makeCluster}} in packages \pkg{parallel} and \pkg{snow}
}
\examples{
# Mixed models:
\dontshow{ if(require(nlme)) \{ }
fm2 <- lme(distance ~ age + Sex, data = Orthodont,
random = ~ 1 | Subject, method = "ML")
ms2 <- dredge(fm2)
# Get top-most models, but fitted by REML:
(confset.d4 <- get.models(ms2, subset = delta < 4, method = "REML"))
\dontrun{
# Get the top model:
get.models(ms2, subset = 1)[[1]]
}
\dontshow{ \} }
}
\keyword{models}