The Research On The Probability Of Default Measure On The Basis Of Logistic Model

In the credit risks management of the Commercial bank, probability of default refers to the possibility that the borrowers who are not able to repay the principal and interest of the bank loan or fulfill the related obligations in a future certain time according to the contract requirement.The probability of default measure to the borrowes which has been listed as the key element in the New Basel Capital Accord's IRB,is an important aspect in the credit risk management of modern commercial bank.As a mainstream method of estimating the probability of default, logistic regression analytic method is not only convenient and flexible, moreover many of its premises and assumptions that are much more in line with the economic reality and the financial data distributed rule,which makes the analysis result of the model to be quite objective. In this paper, in view of questions in the general Logistic default rates model such as the loss of the original data information,multiple collinearities as well as no consideration of time factor, to bring the Logistic default rates reckoning model based on the factorial analysis.Through the introduction of factor analysis and time-weighted indicators and other methods of treatment to improve the Logistic default rates reckoning model, and then using the data to carry out the example analysis, demonstrating the feasibility of the method.And the general two-classify logistic model is divided into two status of default or not merely, which is relatively simple.In this paper, combined with the actual situation in China's commercial banks, according to its ex post facto basis after the loan at the borrower's actual ability to repay loans to the quality of the five-category loan classification, and put two -classify logistic model extended to multiple-classify logistic model, to make full use of existing information and thus to get more precise estimates of probability of default.The Commercial bank may apply the probability of default measure in this paper and use the method of the loan-defaults table estimate the final the probability of default. In this way, both the customer financial data and the customer historical default situation is taken into account to probability of default, greatly improving the accuracy of probability of default for corporate customer loan, and the economic capital measurement model of our group can be supported by the data foundation. Finally we propose the some measures on how to use apply the probability of default measure in this paper.