KMV模型运用于中国上市公司财务困境预警的实证检验  

Testing KMV on the Financial Distress of Listed Companies in China

作　　者：马若微[1] 

机构地区：[1]北京大学经济学院博士后流动站

出　　处：《数理统计与管理》2006年第5期593-601,共9页Journal of Applied Statistics and Management

基　　金：国家自然科学基金项目(项目号70171005)

摘　　要：我国目前关于上市公司财务困境预警的研究大多建立在试误的基础上,缺少经济理论依据;并且经我们证明,按照配对原则设定的分析样本会出现过度抽样等问题,这将会大大高估模型的预警能力。因此本文将基于期权定价理论的KMV模型首次运用到财务困境预警中,引入功率曲线进行对照分析,经过大量的实证研究后得出结论:KMV模型运用到中国上市公司财务困境预警中是完全可行的,而且相对基于大量历史数据得到了Logistic、Fisher等模型,其优势是明显的。So far, many researches on the early - warning of listed companies＇financial distresses based on＇ try and error＇,lacking of economics foundations;and through our testing the pairwise experiments, most models shows the better predictive power. But in practical, they are nothing. So we introduce the KMV model based on the option pricing model to listed companies＇early-warning of financial distress and parallel the discriminant power through power curve Through repeatedly testing, we conclude： a. it is feasible using KMV on the early-warning of listed companies＇financial distresses; b. comparing with Fisher＇s discriminant model and Logistic regression model, KMV is a preferred choice.

关 键 词：KMV 财务困境 预警 

分 类 号：O212[理学—概率论与数理统计]