Variable selection in discriminant analysis in the presence of outliers

Abstract

A n important problem in discriminant analysis that has received little attention in the literature, is the effect of outliers when variable selection is the first step in the anabsis.In this paper we consider two methodi of dealing with this problem. n e classification performance of these methods, and the classrfication performance If possible outliers are ignored, are compared in a simulation study. In the configrations srudied, weJind that none of the approaches consistently ourperforms the others.