DEA-discriminant analysis in the view of goal programming

This article identifies differences and similarities between DEA (Data Envelopment Analysis) and DB (Discriminant Analysis) in the view of GP (Coal Programming). Based upon such characterization, this article proposes a new type of DA technique, referred to as "DEA-Discriminant Analysis (DEA-DA)", that incorporates a methodological strength of DEA into the DA formulation. This research applies the proposed DEA-DA method to both an illustrative data set and a real case study related to Japanese banks. The importance of DEA-DA is confirmed by comparing it with other DA methods. (C) 1999 Elsevier Science B.V. All rights reserved.