Analysis of Corporate Credit Assessment Based on KMV-CART Decision Tree Approach.

Effective evaluation of corporate credit risk is the key to prevent systemic financial risk. In this paper, 25 financial indicators are selected from the company’s solvency, operating ability, profitability and growth ability to construct a credit evaluation system, and the KMV-CART decision tree model is constructed, and then the validity of the KMV-CART decision tree is empirically examined based on the data of 620 companies. The empirical results show that the overall prediction accuracy based on the KMV-CART decision tree model reaches 87.7%, which is better than 82% of the traditional CART decision tree model and 79.3% of the CHAID decision tree model. It shows that the KMV-CART decision tree model can effectively improve the accuracy of credit risk prediction by adding the non-financial indicator of default distance calculated based on the KMV model to the traditional financial indicator system.