[R] cor < -1

I'm having what looks like a quantization problem with cor on R 2.7.0. I get warnings from cor.test because r < -1. R session output is at the bottom. A simple fix would be to add the appropriate checks to the end of the internal cor code. Oliver > x <- c(0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0

[R] 2.7.0 graphics changes and png devices

Hi, I have a question about the graphics changes in 2.7.0. The example is this bit of code: par(mfcol = c(2, 3)) for(i in 1:6) plot(1:10, 1:10, xlab = "xlab", ylab = "ylab", main = "main") In 2.7.0, if I do this interactively (using windows()), I get bold titles. If I do this through a png dev

Re: [R] 2.7.0 graphics changes and png devices

Apologies for the double-post, but I figured I should be slightly more specific. I'm using the windows version of 2.7.0 and 2.6.2. My calls to png is simply png("2.7.0.png") and png("2.6.2.png"), respectively, followed by the appropriate dev.off(). Oliver On 5/19

[R] pairs, par("plt")

? 2. More generally (for my own understanding), why does the par information not do what I expect? Do I have some fundamental misunderstanding of the arrangement of plotting, figure, display, and margin regions within the device? I'm using R 2.5.1, and this behavior occurs on a fresh R c