CreditMetrics is one of main methodologies that measure credit risk through portfolio appoach. One of the obstacles to utilizing CreditMetrics and applying it for optimal credit risk management is that, for measuring exact value of CreditVaR (the maximun loss during given time interval under given confidential level due to deterioration of credit states of obligors), simulation method must be applied in CreditMetrics, but the execution of simulation demands much time and computing resource.
In this paper, we propose variants of methods for increasing the efficiency of executing simulation in CreditMetrics. These methods are based on the idea that much time can be saved by storing and reusing scenarios generated in the process of simulation instead of generating scenarios every time simulation is executed as in general method. These methods are composed of three steps. In the first step, scenarios are generated by applying CreditMetrics simulation to portfolio. In the next step, every time there is a change in the composition of portfolio, stored scenarios are updated or added to reflect the change. In the last step, when a need for measuring credit risk occurs, the answer is searched and provided using the stored scenarios. By applying these methods, it is possible to increase the performance of simulation measuring credit risk and to manage credit risk by more effective way in CreditMetrics framework.
