[RsR] Mann-Whitney Normal approximation formula in wilcox.test function

Usuario R r@u@er@@p@|n @end|ng |rom gm@||@com
Wed Apr 15 10:56:02 CEST 2009


Hi,

I have found the formula for the Mann-Whitney test when a normal
approximation has to be computed, this is:


NTIES <- table(r)
SIGMA <- sqrt((n.x * n.y/12) * ((n.x + n.y + 1) - sum(NTIES^3 - NTIES)/((n.x
+ n.y) * (n.x + n.y - 1))))

Where:
r <- rank(c(x - mu, y))
n.x <- as.double(length(x))
n.y <- as.double(length(y))


This formula does not match with the ones found in the theory (
http://en.wikipedia.org/wiki/Mann-Whitney_U_test#Normal_approximation )

Does anyone know why?

Thanks a lot.
Regards

	[[alternative HTML version deleted]]




More information about the R-SIG-Robust mailing list