Research Of City Commercial Bank Credit Risk Measure Based On Small And Medium-sized Enterprise Customer Group

The risks of Commercial Banks in accordance with the nature are classified into three major risks:credit risk, market risk and operational risk. With the rapid development of economy, all kinds of financial derivatives have occurred, the financial environment determines the credit risk is one of the most important risk facing the commercial banks. The credit risk management has become an indispensable component of daily management for commercial banks. Scientific credit risk management process should include three processes. They are credit risk identification, credit risk measurement and credit risk control. And credit risk measurement is an effective basis for implementation of credit risk management. So the main object of this paper is the credit risk measurement problems of city commercial banks.City commercial banks have experienced reform, transformation and development since they were founded in 1995. Under the correct guide of regulators, city commercial banks promote the reform of its institutional structure, strengthen credit risk management, seek development space. City commercial banks commit to serve the general public, small and medium-sized enterprises and the local economy. They have made a major contribution to the development of local economy. In the modern society, city commercial bank has become an important part of the banking system. Due to the particularity of business scope, the service object and the relationship with local government, the credit risk management for city commercial banks is different from the domestic large and medium-sized commercial banks.This paper compares three traditional Credit Risk measurement methods with three kinds of modern Credit Risk measurement models. They are respectively linear discriminant analysis, logit regression, Probit regression, Credit Metrics value at risk model, KMV model and Credit Risk+model. Final we think traditional Credit Risk measurement method is still the mainstream choice of measuring credit risk for city commercial banks in the short term. This paper selects 66 small and medium-sized enterprises for logit regression model and principal component discriminant analysis model empirical analysis, including 24 ST and* ST enterprises,42 normal enterprises. The empirical analysis shows that the prediction accuracy of the two models are both more than 90 percents. But type 1 error rate of principal component discriminant analysis model is much higher. So logit regression model is worth popularizing and using in the city commercial bank credit risk measurement. At last, the paper offers constructive comments for the city commercial bank credit risk management. Such as accelerating the development of credit database, recruiting credit risk management personnel and perfecting the internal credit rating system.