信用风险度量、债券违约预测与结构化模型扩展  

The Measuring of Credit Risk, Prediction of Bond Default and Expansion of Structured Model

作　　者：杨世伟[1] 李锦成[2] 

机构地区：[1]对外经济贸易大学国际经济贸易学院,北京100029 [2]中国社科院数量经济与技术经济研究所,北京100732

出　　处：《证券市场导报》2015年第10期41-48,共8页Securities Market Herald

基　　金：国家自然科学基金项目"国际资本流动与宏观审慎性政策研究"(批准号:71303044)

摘　　要：随着信用风险在中国市场的积累，债券违约风险逐渐上升。本文通过KMV等信用风险模型对2013-2014年发行的公司债、企业债及私募债进行分析，估计样本公司的资产价值和波动率，同时进行违约距离测算，并结合我国实际情况将其作为多元probit模型的自变量计算得到违约概率。实证结果表明，公司债的违约风险最大而企业债相对安全，私募债虽然风险较小但方差更大，其中银行间私募债风险最低，而中小企业私募债风险更高，这表明应加强上市公司监管并防范私募债风险，同时引导更多优质企业进入以发挥为中小企业融资的功能。As the credit risks have been accumulated in the Chinese market,this paper analyses the bonds between 2013 and 2014 issued by listed company, corporate and privately company through KMV credit risk models etc,estimates the asset value and volatility of sample firms, computes their distance of default and calculates the probability of default by use of the independent variables in the multivariate probit model combined with the actual situation of our country. The empirical results show that the bonds of listed companies have the maximum default risk, and the bonds of corporate are relatively safe. The risk of private bonds is small but has the larger variance. The private bonds among banks have minimum risks, and the risks of small and medium-sized enterprises are higher. This suggests that we should strengthen the supervision of listed companies and prevent private debt risks, at the same time,guide more quality enterprises to solve the financing problems for the small and medium-sized enterprises.

关 键 词：信用风险 PFM模型 违约距离 多元probit模型 

分 类 号：F830.9[经济管理—金融学]