Okay, I solved it. After using the same arguments for "scales" (in particular:
relation="free"), the plots are perfectly aligned.
Cheers,
Marius
On 2010-08-24, at 20:39 , Marius Hofert wrote:
> Dear expeRts,
>
> I would like to have four plots appearing in one figure. The minimal example
> shows this. However, the four figures are not properly aligned. Why? If I
> comment out the scales=... arguments, then it works, but I would like to use
> this... :-)
>
> Cheers,
>
> Marius
>
>
>
> library(lattice)
> set.seed(1)
>
> ## plot: xyplot, histogram, qqplot, acf
> res.plots <- function(x){
>
> x.max <- max(abs(min(x)),abs(max(x)))
>
> ## xyplot
> x0 <- 1
> x1 <- length(x)
> y0 <- -x.max
> y1 <- x.max
> xy <- xyplot(x~(1:length(x)),xlim=c(x0,x1),ylim=c(y0,y1),
>
> aspect=1,scales=list(relation="free",alternating=c(1,1),tck=c(1,0)))
>
> ## histogram
> x0 <- -x.max
> x1 <- x.max
> h <- histogram(x,xlim=c(x0,x1),aspect=1,type="density",
> scales=list(relation="free",alternating=c(1,1),tck=c(1,0)))
>
> ## qqplot
> y0 <- -x.max
> y1 <- x.max
> qq <-
> qqmath(x,ylim=c(y0,y1),aspect=1,scales=list(tck=c(1,0),y=list(rot=90)),
> panel=function(...){
> panel.qqmath(...)
> panel.qqmathline(x)
> })
>
> ## ACF
> myacf <- acf(x,plot=F,lag.max=24)
> x0=0
> x1=24
> y0=-1
> y1=1
> ACF <- xyplot(0~0,xlim=c(x0,x1),ylim=c(y0,y1),aspect=1,
> scales=list(tck=c(1,0),y=list(rot=90)),
> panel=function(...){
> for(i in 1:24){
> panel.segments(i,0,i,myacf$acf[i])
> }
> panel.abline(h=0)
> })
>
> ## plot function
> plot.list <- list(xy,qq,h,ACF)
> counter <- 1
> trellis.device(pdf,paper="special",width=6,height=6,file="myplot.pdf")
> for(i in 1:2){
> for(j in 1:2){
> if(i==2 && j==2) print(plot.list[[counter]], split=c(j,i,2,2))
> else print(plot.list[[counter]], split=c(j,i,2,2), more=TRUE)
> counter <- counter + 1
> }
> }
> dev.off()
>
> }
>
>
> x <- rexp(1000)
> res.plots(x)
>
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
