基于最大熵原理的条件回收率建模分析  

Analysis on the Modeling Method of Conditional Recovery Rate:Based on the Maximum Entropy Principle

作　　者：林洁[1] 

机构地区：[1]厦门大学经济学院,福建厦门361005

出　　处：《统计与信息论坛》2008年第9期32-35,共4页Journal of Statistics and Information

摘　　要：通常使用的数据拟合只是对回收率分布的一个描述,为有效对回收率进行研究,建模时将抵押担保和企业的信用等级这两个因素加以考虑,应用最大熵原理,对违约损失率估计出最佳的条件概率密度。利用该模型不仅能够估计出违约损失率的均值和方差,还可以得出违约损失率的分布密度;另外,该模型也具有更明确的经济学意义。返回检验表明,该模型的估计效果优于单因素模型、动态多元回归模型以及非参数核密度估计法。Data fitting is only a description to recovery rate distribution. Giving consideration to the factors of mortgage guarantee and credit level of enterprise, the author introduces the modeling method of conditional recovery rate. Applying the maximum entropy principle , the author estimates an optimal conditional probability density of the Loss Given Default（LGD）. Not only the mean and variance of LGD but also the distribution density of LGD can be estimated in this model. Furthermore, this model has specific economics meaning. Back -track test shows that this model has better estimation effect to bimodal distribution.

关 键 词：条件回收率 违约损失率 最大熵原理 

分 类 号：F832.4[经济管理—金融学]