EFFICIENT MONTE CARLO METHODS FOR CONVEX RISK MEASURES IN PORTFOLIO CREDIT RISK MODELS

摘要

We discuss efficient Monte Carlo (MC) methods for the estimation of convex risk measures within the portfolio credit risk model CreditMetrics. Our focus lies on the Utility-based Shortfall Risk (SR) measures, as these avoid several deficiencies of the current industry standard Value-at-Risk (VaR). It is demonstrated that the importance sampling method exponential twisting provides computationally efficient SR estimators. Numerical simulations of test portfolios illustrate the good performance of the proposed algorithms.