[Analogue-commits] r384 - pkg/man
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Wed Oct 9 06:30:57 CEST 2013
Author: gsimpson
Date: 2013-10-09 06:30:57 +0200 (Wed, 09 Oct 2013)
New Revision: 384
Added:
pkg/man/predict.pcr.Rd
Log:
document predict.pcr
Added: pkg/man/predict.pcr.Rd
===================================================================
--- pkg/man/predict.pcr.Rd (rev 0)
+++ pkg/man/predict.pcr.Rd 2013-10-09 04:30:57 UTC (rev 384)
@@ -0,0 +1,69 @@
+\name{predict.pcr}
+\alias{predict.pcr}
+
+\title{Predicted values from a principal components regression}
+
+\description{
+ Calculates predicted values from a fitted principal components
+ regression model. Leave-one-out, bootstrap of n-fold crossvalidated
+ predictions are also intended (but not yet implemented).
+}
+
+\usage{
+\method{predict}{pcr}(object, newdata, ncomp = seq_along(object$ncomp),
+ CV = c("none", "LOO", "bootstrap", "nfold"),
+ verbose = FALSE, nboot = 100, nfold = 5, \dots)
+}
+
+\arguments{
+ \item{object}{a fitted model of class \code{"pcr"}, the result of a
+ call to \code{\link{pcr}}.}
+ \item{newdata}{data frame of new observations for which predictions
+ are sought.}
+ \item{ncomp}{numeric; the PCR components for which predictions are
+ sought. Can be a vector in which case predictions for multiple
+ components are computed.}
+ \item{CV}{character; the type of crossvalidation required. Currently,
+ no crossvalidation methods are implemented.}
+ \item{verbose}{logical; should progress on crossvalidation be printed
+ to the console?}
+ \item{nboot}{numeric; the number of bootstrap samples to draw, or in
+ the case of \code{CV = "nfold"} the number of repeats of n-fold CV
+ to perform.}
+ \item{nfold}{numeric; the number of folds to split data into.}
+ \item{\dots}{arguments passed to other methods.}
+}
+
+\details{
+ \code{predict.pcr} arranges for any transformation applied to the
+ training data to be applied to the \code{newdata} prior to
+ prediction.
+}
+
+\value{
+ A matrix of predicted values with rows representing samples in
+ \code{newdata} and columns, the PCR components requested via
+ \code{ncomp}.
+}
+
+\author{Gavin L. Simpson}
+
+\seealso{\code{\link{pcr}}}
+
+\examples{
+## Load the Imbrie & Kipp data and
+## summer sea-surface temperatures
+data(ImbrieKipp)
+data(SumSST)
+
+## choose 10 samples to act as a test set, for illustration
+take <- c(5,58,31,51,42,28,30,57,8,50)
+
+## normal interface and apply Hellinger transformation
+mod <- pcr(ImbrieKipp[-take, ], SumSST[-take], tranFun = Hellinger)
+
+## predictions
+predict(mod, ImbrieKipp[take, ], ncomp = 1:4)
+
+}
+\keyword{methods}
\ No newline at end of file
More information about the Analogue-commits
mailing list