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, 0, 0)
> y <- c(1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1)
> cor.test(x, y)
Pearson's product-moment correlation
data: x and y
t = NaN, df = 13, p-value = NA
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
NaN NaN
sample estimates:
cor
-1
Warning messages:
1: In sqrt(1 - r^2) : NaNs produced
2: In atanh(r) : NaNs produced> cor(x, y) < -1
[1] TRUE> cor(x, y) + 1
[1] -2.220446e-16> .Internal(cor(x, y, 1, FALSE)) < -1
[1] TRUE> .Internal(cor(x, y, 3, FALSE)) < -1
[1] FALSE