基于马尔可夫方法的违约传染平均域模型  

Mean-Field Model of Default Contagion Based on theMarkovian Approach

作　　者：刘久彪[1] LIU Jiu-biao(School of Economics, Tianjin University of Finance ＆ Economics, Tianjin 300222, Chin)

机构地区：[1]天津财经大学经济学院,天津河西300222

出　　处：《经济数学》2017年第2期89-94,共6页Journal of Quantitative Economics

基　　金：教育部人文社会科学研究青年基金项目(12YJC790116)

摘　　要：基于平均域模型,将信用组合分为几个同质的子组合,假设组合中各公司的违约强度不仅取决于宏观经济状况,而且依赖于这些子组合中违约公司的数目,以刻画不同公司间的违约相互作用;并据此建模组合违约过程为连续时间马尔可夫链,借助Kolmogorov微分方程求解信用组合损失分布;最后,通过实例计算分析传染现象对组合损失分布、风险量度的影响.The credit portfolio was divided into several homogeneous sub-portfolio by mean-field model, and the company＇s default intensity was assumed to depend on macroeconomic conditions and the number of defaulted companies in the subportfolio in order to characterize the interaction among different companies.Then, the portfolio default process was modeled as a continuous-time Markov chain, and credit portfolio loss distribution was solved by Kolmogorov differential equation.Finally, we analyzed the influence of default contagion on the portfolio loss distribution and risk measure by the examples.

关 键 词：金融管理学 信用组合 马尔可夫方法 违约传染 平均域模型 

分 类 号：F830[经济管理—金融学]