Dear List,
I sent a related message yesterday, but did not receive it through the
mailing list; on the R-help archives it reads as "An embedded and
charset-unspecified text was scrubbed...". So here it is again, with the
(little) progress I have made since then. Any help would be greatly
appreciated!
I am working with a relatively large correlation matrix (~1600*1600), which
I am looking to plot with function cor.plot in package psych.
cor.plot adds the scalebar as an extra column to the matrix plot, and then
adds an axis with as many tick marks/labels as there are rows in the matrix.
This makes the values unreadable even for relatively small matrices (e.g.
100*100).
In order to make the output more readable, one would want the scalebar to
have a reasonable number of tick marks/labels, say seq(from = zlim[1], to =
zlim[2], by = 0.1) for a given zlim, e.g. zlim=c(-1,1).
So for matrix r, I have modified the relevant bit in cor.plot to
...
nf <- dim(r)[2]
nvar <- dim(r)[1]
...
if (show.legend) {
# at1 <- (0:(nf))/(nf)
at2 <- seq(from = zlim[1], to = zlim[2], by = 0.1)
abline(v = (nf - 0.5)/nf)
axis(4, at = at2, labels = at2, las = 2, ...)
}
However, only values between 0 and 1 are used on the axis label. I believe
this has to do with the analogous behavior in image(), for example:
r <- cor(matrix(rnorm(600), 60, 100))
image(r, zlim=c(-1,1), axes=FALSE)
at2 <- seq(from = -1, to = 1, by = 0.1)
axis(4, at = at2, labels = at2, las = 2, cex.axis=.5)
It is not clear to me from the image() help file why this would be. Could
someone provide any hints on this, or on how to modify cor.plot()
successfully?
Many thanks in advance,
Lara
[[alternative HTML version deleted]]
______________________________________________
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.
