Some Parallels Between Predictive Discriminant Analysis and Multiple Regression

Abstract

The purpose of this paper is to outline some important similarities in, and differences between, predictive discriminant analysis (DA) and multiple regression (MR). The areas covered, chosen for their importance and need for clarification, are estimates of model accuracy, hypothesis testing, and non-least squares models. Some of the parallels are well known, and some appear to have not yet been considered at all.