On 1/26/06, Dieter Menne <[EMAIL PROTECTED]> wrote:
> Dear Latticers,
>
> I want to position two trellis plots of different forms
> side-by-side. The plot types are slightly different,
> aspect=1 required, but panels should look the same in
> both plots. Current workaround uses a guessed factor.
> Any way to improve this?
>
> Dieter
>
>
> library(lattice)
> n1 = 20
> # I cannot rbind df1 and df2, because the x-dimensions are
> # different and must be scaled individually
If that's the only reason, I would suggest rbind-ing them and then use
scales = list(x = "free")
If you want the first two columns to have the same x-limits, you can specify
xlim = list(c(0,25), c(0,25), c(0, 2500))
etc. Otherwise, if you want the panels to have the same physical
dimensions, look at the panel.width and panel.height arguments in
?print.trellis. I can't think of any other way of guaranteeing it.
Deepayan
> df1 = data.frame(y=rnorm(4*n1),x=rep(1:n1,4),
> facA=rep(c("A","B"),each=2*n1),facB=rep(c("C","D"),each=n1))
>
> # I add a dummy facB here, to make sure panels have same structure
> df2 = data.frame(y=rnorm(2*n1),z=rep(100*(1:n1),2),
> facA=rep(c("A","B"),each=n1),facB="C")
>
> # Note: aspect = 1 is required
> p1 = xyplot(y~x|facA*facB,data=df1,main="Plot1",aspect=1,
> between=list(x=2))
> p2 = xyplot(y~z|facA*facB,data=df2,main="Plot2",aspect=1,layout=c(1,2))
>
> wi = 0.61 # this is trial-and-error
> print(p1,position=c(0,0,wi,1),more=T)
> print(p2,position=c(wi,0,1,1),more=F)
______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
