Dear list,
When I try to run the Ansari-Bradley test on two long vectors I obtain a
warning and the p-value is NA:
> set.seed(12)
> x <- rnorm(100000)
> y <- rnorm(100000)
> ansari.test(x,y)
Ansari-Bradley test
data: x and y
AB = 5002890779, p-value = NA
alternative hypothesis: true ratio of scales is not equal to 1
Warning message:
In m * n : NAs produced by integer overflow
This operation occurs inside a function named normalize():
normalize <- function(s, r, TIES, m = length(x), n = length(y)) {
....
}
If I add these two lines to the beginning of the function:
m <- as.double(m)
n <- as.double(n)
Then I obtain the following:
> Myansari.test(x,y)
Ansari-Bradley test
data: x and y
AB = 5002890779, p-value = 0.6599
alternative hypothesis: true ratio of scales is not equal to 1
Is this a proper fix for the problem?
PS: I'm running 2.13.1
Cheers,
Oscar
Oscar M. Rueda, PhD.
Postdoctoral Research Fellow, Breast Cancer Functional Genomics.
Cancer Research UK Cambridge Research Institute.
Li Ka Shing Centre, Robinson Way.
Cambridge CB2 0RE
England
NOTICE AND DISCLAIMER
This e-mail (including any attachments) is intended for ...{{dropped:16}}