跳跃-扩散模型框架下基于鞍点近似的信用组合一致性风险度量

Coherent measures of credit portfolio risk under jump-diffusion framework based on saddlepoint approximations

在债务人资产价值的跳跃-扩散模型框架下,应用鞍点近似方法计算了信用组合损失的概率分布,并以算例具体说明了计算过程,得出了信用组合风险的VaR量度和一致性量度ES.然后,将鞍点近似方法得出的损失分布与蒙特卡罗模拟得出的结果相比较,得出了对于要求精度较高、资产数目较多的信用组合,基于鞍点近似方法计算一致性风险量度是信用风险管理的合适方法的结论.

Based on the obligors assets＇ jump-diffnsion model, this paper calculates credit portfolio loss distribution by applying saddlepoint approximations, and gives an example to show the process, then gets credit portfolio risk measure VaR and coherent risk measure ES. Through comparing the results from saddlepoint approximations and Monte Carlo simulation, we draw the conclusion that it is a proper method to calculates the coherent risk measure by saddlepoint approximations for highly accurate credit portfolio risk management.