个人抵押贷款违约的相关性因素--基于Copula模型的研究  

The Related Factors of the Defaults of Individual Mortgage Loans——A Study Based on Copula Model

作　　者：王鹰翔[1,2] 王瑞峰 

机构地区：[1]西安交通大学管理学院 [2]中国工商银行信贷管理部

出　　处：《金融论坛》2010年第11期10-14,共5页Finance Forum

摘　　要：运用国内商业银行积累的大量数据,统计得到银行个人客户住房抵押贷款多年度、不同信用等级、不同身份特征、分行业和分地区的违约情况,进行非线性的拟合分析,并采用Copula函数度量个人客户违约之间的相关性及厚尾特征。研究表明,房屋价格、客户性别以及受教育程度等与违约概率相关性比较低,在考察的样本区间内,这些因素不显著导致违约发生。另外,信用等级、收入结构和抵押担保剩余额度是影响个人违约决策的重要变量。所采用的模型在个人住房抵押贷款定价与风险管理中获得较好效果,银行可以根据违约状况的变动制定动态利率,随时准备弥补损失。By means of domestic commercial banks＇ accumulated data, the author counts the defaults of the banks＇ individual mortgage loans of the multi-year, different credit ratings, different identities, different sectors and different regions, makes the nonlinear simulation and examines the related factors of the individual defaults and the fat tail with the Copula function. The results show that there is a relatively low correlation between the housing price, the customer＇s gender and education level and the probability of default. Within the study samples, these factors do not significantly cause the default. In addition, the credit rating, the income structure and the remaining amount of the collateral are important variables in the individual decision-making of the default. The model used in the paper obtains better results in the pricing and risk management of the individual housing mortgage loans. According to the change of the default, the bank can determine dynamic interest rates and get ready to make up for losses whenever necessary.

关 键 词：商业银行 个人住房抵押贷款 违约风险 COPULA函数 模拟测算 

分 类 号：F830.5[经济管理—金融学]