I am using R corrplot library. It looks amazing, however to produce a really good plot I want to change the labels of rows and columns of the correlation matrix.
One solution is to do something like this:
cbak <- colnames(my.data.frame)
colnames(my.data.frame) <- c("a", "set", "of", "labels")
corrplot(cor(my.data.frame))
colnames(my.data.frame) <- cbak
However this looks strange and ugly.
I guess I should work with the labels parameter of a text() function, but I can't figure how.
corrplot(cor(my.data.frame), labels=c("a", "set", "of", "labels"))
results in
Error in text.default(pos.xlabel[, 1], pos.xlabel[, 2], newcolnames, srt = tl.srt, :
invalid 'pos' value
In addition: Warning message:
In text.default(pos.xlabel[, 1], pos.xlabel[, 2], newcolnames, srt = tl.srt, :
NAs introduced by coercion
How to do this correctly?
Answer
In the current corrplot version 0.75, you cannot use labels parameter because the X and Y labels are computed within the corrplot() function from colnames() and rownames() of the input corr matrix.
I'm using similar approach as you have suggested:
M <- cor(mtcars)
colnames(M) <- c("a", "set", "of", "x", "labels", 1:6)
corrplot(M, method = "color")
BTW, I linked this stackoverflow question from our github issue tracker: https://github.com/taiyun/corrplot/issues/20
UPDATE:
In the current corrplot version 0.78, also plotmath expressions are allowed in variable names. Just prefix your name with one of the characters ":", "=" or "$".
Example:
M <- cor(mtcars)[1:5,1:5]
colnames(M) <- c("alpha", "beta", ":alpha+beta", ":a[0]", "=a[beta]")
rownames(M) <- c("alpha", "beta", NA, "$a[0]", "$ a[beta]")
corrplot(M)
