How does one compute the null distribution for the Mann-Kendall test of monotone trend? The table in Hollander & Wolfe has lots of gaps in it, and my sample size is too small for a normal approximation.
Thank you.
Valerie
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
