Research on Credit Risk Measurement of China’s Listed Companies with KMV Model

Abstract: This thesis takes 200 Chinese listed companies as examples within ten years from 2009 to 2018, of which 100 are ST companies and the other 100 are non-ST companies. ST company is a company that has financial problems and was then implemented with special treatment by the China Securities Regulatory Commission. The traditional credit risk measurement model, the Z-Score model, is chosen to be compared with the KMV model to test whether the KMV model is more suitable for the Chinese financial market. Through comparative analysis, we can judge whether the KMV, which has been highly praised in recent years, has the apparent ability of identification and prediction for defaulting companies. Then, with the ROC curve, the ability to identify and predict of the two models can be further examined. A more accurate credit risk measurement model can help investors identify the company's financial status and avoid property losses caused by unnecessary risks as much as possible. The experimental results show that, for the Chinese stock market, the differentiation ability of the KMV model is more suitable than that of the Z-Score model, and it is also more suitable as a prediction model. Finally, if the KMV model is to be vigorously promoted in the Chinese market, it must overcome its shortcomings. Therefore, suggestions for improvement are listed in this thesis.