A Multiple-Comparisons Method Based on the Distribution of the Root Node Distance of a Binary Tree

Abstract

This articleproposesan easyto implementcluster-basedmethodforidentifyinggroups of nonhomogeneousmeans. The method overcomes the common problem of the classical multiple-comparisonmethods that lead to the construction of groups that often have substantialoverlap.In addition,it solves the problemof othercluster-basedmethodsthat do not havea known levelof signicanceand are not easy to apply.The new procedureis compared by simulation with a set of classical multiple-comparisonmethods and a cluster-basedone. Results show that the new procedure compares quite favorably with those included in this article.