Assessing ®nancial risks using a multicriteria sorting procedure: the case of country risk assessment

The assessment of ®nancial risks is a problem of major interest for corporate entities (organizations, ®nancial institutions, ®rms, etc.). The vulnerable economic and ®nancial environments necessitate the development of operational approaches to measure and control ®nancial risks. Most of the methodologies that have been proposed in the past employ a probabilistic notion of risk. This paper proposes an alternative approach to measure ®nancial risks, considering their multidimensional nature. The proposed approach is based on the multicriteria decision aid (MCDA) method Multi-Group Hierarchical DIScrimination (M.H.DIS). The aim of the M.H.DIS method within the ®nancial risk assessment context is to develop a set of additive utility functions that classify the considered alternatives (®rms, investment projects, portfolios, countries, etc.) into prede®ned risk classes. The eciency of the method is illustrated through a case study regarding the country risk assessment problem. Using the M.H.DIS method a discrimination model is developed that classi®es the countries into four groups, and measures the corresponding creditworthiness and risk of the countries. Several validation tests are performed in order to compare the classi®cation results obtained through M.H.DIS to the results obtained through multiple discriminant analysis. 7 2000 Elsevier Science Ltd. All rights reserved.