作　　者： ;

机构地区： 华南理工大学

出　　处： 《湖南大学学报（自然科学版）》 2011年第3期87-92,共6页

摘　　要： 违约概率是公司信用风险管理的重要参数,是计算公司预期违约损失、债券定价以及信贷组合管理的基础.为了更准确更有效地测度违约概率,引入了延迟滤波来定义不完全信息,改变信息结构,减少'噪音',以提高违约概率测度的精确性;用重随机Poisson过程的延迟滤波取代由布朗运动所形成的自然滤波的鞅来计算违约概率以避免一些金融资产的运动并不具备鞅性的缺陷;信用分析模型中的难点——违约强度可根据市场信息来估计.这一方法对于缺少历史违约数据的新兴市场公司债违约概率的测度,提供了一种可供借鉴的思路和方法. As an important parameter for the credit risk management,default probability is the basis for the calculation of expected default loss,bond pricing and credit portfolio management.In order to estimate default probability more accurately and effectively,this paper defines the incomplete information with the delayed filtrations to change the information structure and filtrate the noise so as to increase the accuracy of the estimation of default probability and replaces martingale with respect to the natural filtration produced by the Brownian motion with the delayed filtration produced by the doubly stochastic Poisson process to calculate the default probability so as to avoid the limitation that the movement of some financial assets is not a martingale.With this key of the credit model,default intensity can be estimated from market information.The method put forward in this paper can be applied for emerging markets without sufficient historical default data.

关 键 词： 延迟滤波 违约强度 违约概率

领　　域： []