[Lme4-commits] r1585 - pkg/lme4Eigen/man

[noreply at r-forge.r-project.org]

Author: dmbates
Date: 2012-02-09 19:40:00 +0100 (Thu, 09 Feb 2012)
New Revision: 1585

Added:
   pkg/lme4Eigen/man/predict.merMod.Rd
   pkg/lme4Eigen/man/simulate.merMod.Rd
Log:
Add new documentation files.


Added: pkg/lme4Eigen/man/predict.merMod.Rd
===================================================================
--- pkg/lme4Eigen/man/predict.merMod.Rd	                        (rev 0)
+++ pkg/lme4Eigen/man/predict.merMod.Rd	2012-02-09 18:40:00 UTC (rev 1585)
@@ -0,0 +1,58 @@
+\name{predict.merMod}
+\alias{predict.merMod}
+\title{Predictions from a model at new data values}
+\usage{
+  \method{predict}{merMod} (object, newdata = NULL,
+    REform = NULL, terms = NULL,
+    type = c("link", "response"), allow.new.levels = FALSE,
+    ...)
+}
+\arguments{
+  \item{object}{a fitted model object}
+
+  \item{newdata}{data frame for which to evaluate
+  predictions}
+
+  \item{REform}{formula for random effects to include.  If
+  NULL, include all random effects; if NA, include no
+  random effects}
+
+  \item{terms}{a \code{\link{terms}} object - not used at
+  present}
+
+  \item{type}{character string - either \code{"link"}, the
+  default, or \code{"response"} indicating the type of
+  prediction object returned.}
+
+  \item{allow.new.levels}{(logical) if FALSE, then any new
+  levels detected in \code{newdata} will trigger an error;
+  if TRUE, then the prediction will use the unconditional
+  (population-level) values for data with previously
+  unobserved levels}
+
+  \item{...}{optional additional parameters.  None are used
+  at present.}
+}
+\value{
+  a numeric vector of predicted values
+}
+\description{
+  \code{\link{predict}} method for
+  \code{\linkS4class{merMod}} objects
+}
+\note{
+  explain why there is no option for computing standard
+  errors of predictions!
+
+  offsets not yet handled
+}
+\examples{
+(gm1 <- glmer(cbind(incidence, size - incidence) ~ period + (1 |herd), cbpp, binomial))
+str(p0 <- predict(gm1))            # fitted values
+str(p1 <- predict(gm1,REform=NA))  # fitted values, unconditional (level-0)
+newdata <- with(cbpp, expand.grid(period=unique(period), herd=unique(herd)))
+str(p2 <- predict(gm1,newdata))    # new data, all RE
+str(p3 <- predict(gm1,newdata,REform=NA)) # new data, level-0
+str(p4 <- predict(gm1,newdata,REform=~(1|herd))) # explicitly specify RE
+}
+

Added: pkg/lme4Eigen/man/simulate.merMod.Rd
===================================================================
--- pkg/lme4Eigen/man/simulate.merMod.Rd	                        (rev 0)
+++ pkg/lme4Eigen/man/simulate.merMod.Rd	2012-02-09 18:40:00 UTC (rev 1585)
@@ -0,0 +1,39 @@
+\name{simulate.merMod}
+\alias{simulate.merMod}
+\title{Simulate responses from a \code{\linkS4class{merMod}} object}
+\usage{
+  \method{simulate}{merMod} (object, nsim = 1, seed = NULL,
+    use.u = FALSE, ...)
+}
+\arguments{
+  \item{object}{a fitted model object}
+
+  \item{nsim}{positive integer scalar - the number of
+  responses to simulate}
+
+  \item{seed}{an optional seed to be used in
+  \code{set.seed} immediately before the simulation so as
+  to to generate a reproducible sample.}
+
+  \item{use.u}{(logical) generate new random-effects values
+  (FALSE) or generate a simulation condition on the current
+  random-effects estimates (TRUE)?}
+
+  \item{...}{optional additional arguments, none are used
+  at present}
+}
+\description{
+  Simulate responses from the model represented by a fitted
+  model object
+}
+\examples{
+## test whether fitted models are consistent with the
+##  observed number of zeros in CBPP data set:
+gm1 <- glmer(cbind(incidence, size - incidence) ~ period + (1 | herd),
+             data = cbpp, family = binomial)
+gg <- simulate(gm1,1000)
+zeros <- sapply(gg,function(x) sum(x[,"incidence"]==0))
+plot(table(zeros))
+abline(v=sum(cbpp$incidence==0),col=2)
+}
+