The following function is designed to work with a logit link. It can

easily be generalized to work with any link. The SEs and CIs are

evaluated accounting for all sources of random variation. The plot

may not be much help unless there is just one explanatory variate.

`ciplot` <-

function(obj=glmm2, data=data.site, xcol=2, nam="litter"){

cilim <- function(obj, xcol){

b <- fixef(obj)

vcov <- summary(obj)@vcov

X <- unique(model.matrix(obj))

hat <- X%*%b

pval <- exp(hat)/(1+exp(hat)) # NB, designed for logit link

U <- chol(as.matrix(summary(obj)@vcov))

se <- sqrt(apply(X%*%t(U), 1, function(x)sum(x^2)))

list(hat=hat, se=se, x=X[,xcol])

}

limfo <- cilim(obj, xcol)

hat <- limfo$hat

se <- limfo$se

x <- limfo$x

upper <- hat+2*se

lower <- hat-2*se

ord <- order(x)

plot(x, hat, yaxt="n", type="l", xlab=nam, ylab="")

rug(x)

lines(x[ord], lower[ord])

lines(x[ord], upper[ord])

ploc <- c(0.01, 0.05, 0.1, 0.2, 0.5, 0.8, 0.9)

axis(2, at=log(ploc/(1-ploc)), labels=paste(ploc), las=2)

}

## Usage

glmm2 <- lmer(rcr ~ litter + (1 | Farm), family=binomial,

data=data.site)

ciplot(obj=glmm2)

John Maindonald email:

john.maindonald@anu.edu.auphone : +61 2 (6125)3473 fax : +61 2(6125)5549

Centre for Mathematics & Its Applications, Room 1194,

John Dedman Mathematical Sciences Building (Building 27)

Australian National University, Canberra ACT 0200.

On 8 May 2008, at 8:00 PM, r-help-request@r-project.org wrote:

From: "May, Roel" <roel.may@nina.no>

Date: 8 May 2008 12:23:15 AM

To: r-help@r-project.org

Subject: [R] predict lmer

Hi,

I am using lmer to analyze habitat selection in wolverines using the

following model:

(me.fit.of <-

lmer(USED~1+STEP+ALT+ALT2+relM+relM:ALT+(1|ID)+(1|

ID:TRKPT2),data=vdata,

control=list(usePQL=TRUE),family=poisson,method="Laplace"))

Here, the habitat selection is calaculated using a so-called discrete

choice model where each used location has a certain number of

alternatives which the animal could have chosen. These sets of

locations

are captured using the TRKPT2 random grouping. However, these sets are

also clustered over the different individuals (ID). USED is my binary

dependent variable which is 1 for used locations and zero for unused

locations. The other are my predictors.

I would like to predict the model fit at different values of the

predictors, but does anyone know whether it is possible to do this? I

have looked around at the R-sites and in help but it seems that there

doesn't exist a predict function for lmer???

I hope someone can help me with this; point me to the right

functions or

tell me to just forget it....

Thanks in advance!

Cheers Roel

Roel May

Norwegian Institute for Nature Research

Tungasletta 2, NO-7089 Trondheim, Norway