Empirical Research On Cohesive Value At Risk Model In Credit Portfolio Measurement

This paper focuses on the measurement of credit risk and the management of credit risk portfolio. The measurement and management of credit risk have been gaining more and more attentions in recent years. The traditional approaches such as subjective analysis and accounting based credit-scoring systems are no longer sufficient for credit risk management because of the emergence of more sophisticated credit products and derivatives products , the global capital market and the increasing competition in the credit market. However, the most important driving force is the more and more concentration risk. The response to these forces academics and practitioners alike have responded by: (I) moved away from only analyzing the credit risk of individual loans and securities towards developing measures of credit portfolio risk; (II) developing new and more sophisticated models to price credit risk. The contribution of this study is three-fold.In the first part, we trace development in the credit risk measurement literature over the last 20 year, and find that the traditional individual approaches are no longer sufficient for nowadays credit management. In order to respond the change, we bring inthe portfolio way-Value at Risk model. VaR models have many advantage ofmeasure portfolio risk and today some advanced bank are used it for their risk management. In this study, we introduce these popular Credit-VaR models and find someserious limitations.In the second part, for skewed return-loss distributions of credit capital and the lack of sub-additivity of VaR, we examine a new approach for credit optimization. The model is based on the Cohesive Value at Risk (CVaR) risk measure, the expected loss exceeding Value at Risk. This model can simultaneously adjust all positions in a portfolio of financial instruments in order to minimize CVaR subject to trading and return constrains. The credit optimization problem is solved effectively by linear programming. In the last part, we demonstrate the approach with a portfolio of china bond market of shanghai to verify the efficiency of the algorithm, and make some advice to China's credit risk measurement and management.