Is there something like a pairwise version of the wilcoxon-mann-whitney test, i.e. the version of the pairwise Wilcoxon signed rank test, but with alternative hypothesis?
I don't seem to be able to find any information on such a thing, neither on the web, nor in hypothesis testing handbooks.
Maybe there is no special paired version of the wmw and even if the data is paired, using wmw is fine...
With pairwise data, you are typically testing hypotheses about the differences, correct? If so, the Wilcoxon signed rank test on the pairwise differences is a good test to use.