A general nonparametric methodology for performing multiple comparisons of treatment effects in r-way mixed models is presented. Comparisons can be made among all treatments or for several treatments against a control. The methods can be used on either the raw data or ranked data. Essentially, we perform M permutations of the observations, consistent with our experimental design. For the comparison of all treatments, any "treatment differences" that are larger than 95 percent of the "simulated ranges" are declared significantly different. Similar methods are also developed for one-sided treatments versus control multiple comparison.
No takes yet. Share an insight, caveat, or question.
Shuster et al. (1979) studied this question.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: