I am going through the taylor.diagram() function of the plotrix package, in order to modify it to meet my needs.
Oddly, I came upon a line of code that is intriguing me:
S <- (2 * (1 + R))/(sd.f + (1/sd.f))^2
This appears at the end of the else part of the "pos.cor if".
I suspect this has to do with some shift to the position of the points to be drawn afterwards, given that if pos.cor is false then the whole range of correlations is to be plotted (from -1 to 1), and thus the plot region becomes modified.
However, the strange part is that S is found nowhere else in taylor.diagram !!!
Am I missing something (and S is being used in some cryptic way), or is this a bug?
Cheers