Discriminant Analysis

Discriminant analysis is used in situations where the clusters are known a priori. The aim of discriminant analysis is to classify an observation, or several observations, into these known groups. For instance, in credit scoring, a bank knows from past experience that there are good customers (who repay their loan without any problems) and bad customers (who have had difficulties repaying their loans). When a new customer asks for a loan, the bank has to decide whether or not to give the loan. The information of the bank is given in two data sets: multivariate observations on the two categories of customers (including age, salary, marital status, the amount of the loan, and the like).