The Research On Commercial Bank Real Estate Credit Default Probability Based On KMV Model

In recent years,the global skyrocketing property market has made China's real estate bubble more and more large,which has caused China's real estate enterprises to face many severe tests such as unable to repay loans on time.However,real estate enterprises are highly dependent on commercial bank loans and the current backward risk management system of commercial banks in China will inevitably lead to greater real estate credit default risk.Therefore,it is of great practical significance to quantitatively study and manage the real estate credit risk of commercial banks.Based on the analysis of commercial banks,this paper takes the measurement of credit default probability of listed real estate enterprises in China as the research object,and studies the whole article from five parts.This paper first expounds the formation mechanism of real estate credit default risk of banks,and analyzes the current situation of real estate credit default risk of China's commercial banks and the possible multi-faceted problems.Then the basic principle,applicability and calculation method of KMV model are expounded.According to the actual situation of the research,the parameters involved in the KMV model are designed and the default parameters of the default point are corrected.The ratio of long-term liabilities to short-term liabilities in default points measured by default probability is0.65.Finally,15 ST listed companies and 15 matched non-ST companies in2011-2018 were selected as research samples.Based on the revised KMV model and the financial and stock market data of listed real estate companies in 2011-2018,Matlab software was used to calculate default.Distance and expected probability of default.In order to verify the validity and accuracy of the model,we first compare the empirical results from the perspective of statistical analysis,and find that the ST companies have a higher probability of default,indicating that the model can better reflect the credit default status of listed real estate enterprises;The independent sample T test of the empirical results using SPSS software shows that the model has limited ability to identify the difference in credit default risk levels between ST companies and non-ST enterprises.Finally,using the ROC curve to evaluate the accuracy of the model,it is found that the model has a good discriminating ability for the credit default risk of listed real estate enterprises.This paper believes that the model measurement results may be related to the non-validity of China's stock market and the lack of a large historical defaultdatabase.Therefore,the recommendations of commercial banks to prevent real estate credit default risk are put forward from two angles,and the full text,especially the empirical part,is summarized in conclusion.