Means Grouping based on Studentized Midrange

Abstract

The purpose of this paper was to develop two procedures of multiple comparisons based on methods of clustering means, that is, MGM and MGR tests. The first is based on the studentized midrange distribution, and the second based on the studentized range distribution. The tests showed performance equivalent to the performance of the compared tests. Like the tests presented that were based on methods of grouping averages, the MGM and MGR tests did not control the experimentwise error rate for almost all evaluated scenarios. However, under the complete $H_1$ hypothesis, these tests showed high power, with emphasis on the MGM test. Thus, what we propose is yet another test alternative without ambiguity in its results, and not a substitution of the traditional tests already present in the literature.