Research On Credit Risk Quantitative Measurement Of Listed Companies In China

Banks play a decisive role in the development of Financial and economics in the country.The credit-risk, as the most important risk confronted by the commercial banks,not onlyinfluence the safty of commercial banks,but aslo, the stability of financial system likedominoes. Nowdays, studies on the credit-risk quantitative measurement are the appliment ofthe results of foreigners. Let the measurement of credit risk of commercial bank be a researchtopic of this article, by combining traditional methods with modern methods to build a newcredit-risk measurement modle.So that we can narrow the gap with other countries, andimprove China’s banking industry’s whole competitiveness.Through analyzing each credit-risk measurement modle, the article sums up theadvantages and disadvantages and applicability in our contry. Then, bringingDefault-Distance, as a new index into the logistic Regression model, to build a newmeasurement modle. In calculating Default-Distance, we make certain amendments on someof the KMV model parameters. Firstly, we use the net assets per share to calculate the valueof non-tradable shares. Secondly, considering the rate of return’s non-normality, GARCHmodle is being used, instead of traditional statistical methods, to calculate the equity’svolatility. Thirdly, through comparing each Default-Distance’s significance, to choose thebest default-point from the four default-points that seted before.On the base of amendments of KMV modle, using the history of the sample stocktransaction data to calculate the Default-Distance. Then, use the factor analysis to determinethe commaon factor of the financial indicators of listed companies. Based on the aboveresults, we build two modles, according to whether the modle’s independent variables includeDefault-Distance. Through the comparison and analysis on the differentiation effect of twodifferent models, the conclusion is: Default-Distance can enhance the ability of forecastingand interpreting of prediction model evidently. The combining of dynamic nature of theKMV model with Logistic regression model can measure the credit risk of commercial banksin China better, and improve China’s commercial banks’ capability to warn credit-risk early.