Permutation Theory in the Derivation of Robust Criteria and the Study of Departures from Assumption

Abstract

SYNOPSIS In the practical circumstances in which statistical procedures are applied, little is usually known of the validity of assumptions such as the normality of the error distribution. Procedures are required which are “robust” (insensitive to changes in extraneous factors not under test) as well as powerful (sensitive to specific factors under test). Permutation theory, which provides one method for deriving robust criteria, is discussed and applied to the problem of comparing variances.