基于Logistic模型的单个信用资产违约概率预测  

Forecasting the Default Probability of Single Credit Assets on the Basis of the Logistic Model

作　　者：白保中[1] 朱世武[1] 

机构地区：[1]清华大学经济管理学院,北京100084

出　　处：《广东金融学院学报》2008年第2期57-65,共9页Journal of Guangdong University of Finance

基　　金：国家自然科学基金项目(70273016)

摘　　要：在信用风险管理领域，由于中国商业银行信贷数据只满足Logistic模型的要求，因而预测单个信用资产违约率只能以Logistic模型为主线建模。以中国某商业银行1999-2005年的信贷数据为样本，实证分析得出，企业本身、宏观经济、地区及行业四方面因素对企业违约概率存在显著相关性。通过以上述四因素为变量，所构建的预测电力、公路、城镇建设三个行业信用资产违约概率的Logistic模型分析与预测单个信用资产违约的结果来看，四要素模型对中国商业银行的信用风险管理具有参照价值。Because the credit data of China＇s commercial banks only meet the requirements of the Logistic model in the field of credit risk management, the default probability of single credit asset can only be forecasted with the Logistic model. The credit data of a commercial bank of China from 1999 through 2005 show that enterprises, the macroeconomy, location and industries exert great influence on the default probability of enterprises. With these four factors as variants, the forecast by the Logistic model for forecasting the default probability of the credit assets in the electric power, highways and urban construction industries shows that the four-variant model has a reference value for the credit risk management of China＇s commercial banks.

关 键 词：违约概率 信用评级 LOGISTIC模型 

分 类 号：F832.4[经济管理—金融学]