A multicriteria hierarchical discrimination approach for credit risk problems

Abstract

Recently, banks and credit institutions have shown an increased interest in developing and implementing credit-scoring systems for taking corporate and consumer credit granting decisions. The objective of such systems is to analyze the characteristics of each applicant (firm or individual) and support the decision making process regarding the acceptance or the rejection of the credit application. This paper addresses this problem through the use of a multicriteria classi - fication technique, the M.H.DIS method (Multi-group Hierarchical DIScrimination). M.H.DIS is applied to real-world case studies regarding the assessment of corporate credit risk and the evaluation of credit card applications. The results obtained through the M.H.DIS method are compared to the results of three wellknown statistical techniques, namely linear and quadratic discriminant analysis, as well as logit analysis.