You didn't do what Prof. Ripley suggested - adding a ... argument.
Here's a crude version of what you want; I'm sure there's a more elegant
solution for passing the needed data to the panel function.
panel.cor <- function(x, y, digits=2, prefix="", splitvar, col.cor, ...)
{
usr <- par("usr"); on.exit(par(usr))
par(usr = c(0, 1, 0, 1))
r <- abs(cor(x, y))
if(!missing(splitvar)) {
r <- c(r, abs(sapply(lapply(split(cbind.data.frame(x, y),
splitvar), cor), function(x)x[1,2])))
}
txt <- format(c(r, 0.123456789), digits=digits)[1:4]
txt <- paste(prefix, txt, sep="")
if(missing(col.cor)) col.cor <- c("black", "red", "green3", "blue")
for(i in 1:length(txt)) {
text(0.5, (1/(length(txt)+1))*i, txt[i], col = col.cor[i])
}
}
pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species",
pch = 21, bg = c("red", "green3", "blue")[unclass(iris$Species)],
lower.panel=panel.cor, splitvar=iris$Species)
On Mon, Jan 5, 2009 at 5:32 PM, herwig <bachmannher...@hotmail.com> wrote:
>
> Dear Prof. Ripley,
>
> thanks for your reply.
> Unfortunately I still was not able to solve the problem.
> I tried it with the Iris data and you can find the code below:
>
>> panel.cor <- function(x, y, digits=2, prefix="", cex.cor)
>
> {
> usr <- par("usr"); on.exit(par(usr))
> par(usr = c(0, 1, 0, 1))
> r <- abs(cor(x, y))
> txt <- format(c(r, 0.123456789), digits=digits)[1]
> txt <- paste(prefix, txt, sep="")
> if(missing(cex.cor)) cex.cor <- 0.8/strwidth(txt)
> text(0.5, 0.5, txt, cex = cex.cor * r)
> }
>
> pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species",
> pch = 21, bg = c("red", "green3", "blue")[unclass(iris$Species)],
> lower.panel=panel.cor )
>
> What I would ideally like to have is that the upper panel shows the plot
> with the iris$Species color coded and the lower panel showing the
> correlation coefficients for the total data in the plot, but also for the
> each iris$Species with the same color code as in the upper panels.
> So the lower panel should therefore actually show 4 correlation coefficients
> in each panel (color coded); It does not need to be sized according to the
> value of the correlation coefficient as shown in the example above.
> Any help with this would be appreciated a lot,
>
> Kind regards,
>
> Herwig
>
>
>
--
Sarah Goslee
http://www.functionaldiversity.org
______________________________________________
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.
