Study On The Credit Risk Of The Listed Small Enterprises Which Has Based On The Models KMV Vine-Copula

As an important component of the financial sector, the healthy and orderly development of commercial banks effect the whole nation’s economy. Besides, the credit risk is the main risks faced by commercial banks. Therefore, the effective risk measurement and management becomes the important subject faced by commercial banks.The paper tries to have a research on enterprise credit risks from the angle of unidimension and multi-dimension. First of all, the paper introduces the basic theory of KMV model and the model structure; then it makes empirical research on the credit risk of single assets with the modified KMV model. Based on analyzing the influential factors of the value of equity, the paper uses non-tradable shares and tradable shares to calculate the value of equity with using the logarithm yield of equity value directly to analyze wave characteristics. Due to the large fluctuations follow with big, the paper build GARCH model to depict the asymmetry of the financial data which general GARCH model cannot do; What’s more, on multidimensional perspective, the paper introduces cany copulas, combined with the modified KMV model, to analyze the four companies’ probability of default based on the same plate joint. The conclusion is it is interplay ed between the companies, and then based on the modified KMV-copulas model, I probe into the effect degree when the corporate whose benefit is relatively good but has default joint probability on poor corporate vice versa. The result is good corporate impact profounder than the poor corporate. Besides, I also attempt to join the Pair Copula under vine structure to describe high-dimensional related structure to extend credit risk research from two-dimensional state to high dimension. Pair copula not only retains its advantage, and it is also more flexible and less constrains compared with multiple copula function in the construction of high dimensional dependency structure. In this paper, to fit the actual data distribution with binary Guassion copula, T copula, Clayton copula and Gumbel copula in copula function choosing to get more accurate results in measuring commercial bank credit risk based on the selected copula according to the AIC criterion; Finally, I put forward the suggestions and expectations based on the empirical results from the paper and current social situation.