Research On Credit Risk Measurement Method Of Listed Private Enterprises

Since the substantial default of the "11-surplus Japanese debt" in March 2014,China's corporate bond default has kicked off.According to the Wind database,as of December 31,2019,there were a total of 425 defaults on corporate credit bonds in China.As credit risk develops towards diversification and complexity,corporate credit risk issues attract wide attention from regulatory authorities and domestic banks and other financial institutions.Private enterprises account for a considerable proportion in the industrial structure.Obviously,they also occupy an important position in the national economy.And their operation and management will directly affect the overall development of our economy.Under this background,it is of high theoretical and practical significance to deeply study the credit risk of listed companies in China's private enterprises and to choose an appropriate evaluation model and indicator system to measure the credit risk of enterprises more scientifically.The content of this study and the conclusions drawn are as follows:(1)The company's 2018 fundamental data of 360 private companies was selected,including 60 listed companies with bond defaults and special treatment(ST)and 300 listed companies with no bond defaults.Corrected the equity value volatility through GARCH(1,1)and then substituted it into the KMV model to solve the default distance.The results showed that the default distance in KMV has a good ability to identify credit risk,and the KMV model corrected by GARCH(1,1)has a better credit evaluation effect.(2)In order to obtain a more effective credit risk evaluation model,20 financial indicators and 3 non-financial indicators of the enterprise were selected.Then the five factors of credit risk evaluation were constructed by the principal component analysis method,which can represent the solvency,operating ability,profitability and growth ability.Using these factors,a Logistic regression credit evaluation model was established.The results showed that all factors have significant effects in the model and the model has good fitting and predictability.Logistic regression is a reliable measure model of credit risk.(3)On the basis of the logistic regression model,the index system added the default distance DD as a new explanatory variable.Then a KMV-Logistic mixed model was established.Next compared and analyzed the Logistic model and the KMV-Logistic model.On the one hand,the results showed that the modified KMV-Logictic mixed model has a decision coefficient which can reach 87.9%.On the other hand,compared to the Logistic regression model,the TPR index of the modified KMV-Logictic mixed model increased by 6.7%,the FPR index of which decreased by 4.0%and the overall accuracy rate of which has been improved.4.4%.It is more effective to use the revised KMV-Logistic model for early warning of credit risk of private enterprises.Through GARCH(1,1)modified KMV model,Logistic model and modified KMV-Logistic mixed model,the empirical research on the credit risk of private listed companies was conducted.And the model optimization and model application were summarized.This article provides suggestions for the measurement of credit risk of private enterprises in practice,which has certain reference value for the selection of credit risk measurement methods of private enterprises.