The Research Of Credit Risk Assessment Of Small-and-Medium-sized Enterprises Based On Logistic Regression Model

Along with the financial market being more and more open, effective riskmanagement mechanism of commercial banks becomes more and more important inour country. Now, the number of commercial banks in our country is increasing, andthe competition between Banks is also more and more fierce. It’s really a veryimportant problem need to be solved that how to expand business and minimize therisk effectively at the same time. At present, according to the statistics, small andmedium-sized (SMS) enterprises take more than ninety-nine percent of the totalnumber of enterprise, which doubtless provide a large business market for commercialbanks. But banks are unwilling to lend to SMS enterprises because SMS enterprises aresmall scale, weak ability to resist risk, short cycle, bad information transparency. Ifthere is a reasonable and effective credit risk evaluation system, banks can makeeffective judgment degree of SMS enterprise, which can eliminate doubts when bankslend to SMS enterprise, than we can see a "win-win" situation between banks and SMSenterprises.In this paper, we analyzed and compared the methods of existent credit riskmeasurement, then pointed out that Logistic Regression method was the most suitableone to measure the credit risk of SMS enterprises in our country. Based on the LogisticRegression Model, we analyzed the financial data and material in2010and2011of120SME enterprises. Result showed that Logistic regression model was feasible tomeasure the credit risk of SMS enterprises in our country, and the result ofmeasurement was accurate. The paper was divided into five parts:In part one, we briefly introduced research background and significance, thenintroduce the domestic and foreign literature about methods of existent credit riskmeasurement, finally introduced the ideal, frame, research method and innovativepoints of the paper. In part two, we introduced credit evaluation theory, including thedefinition of credit risk, the concept of credit risk assessment and all the credit risk evaluation methods at present. Then analyzed and compared the methods of existentcredit risk measurement, pointed out that Logistic Regression method was the mostsuitable one to measure the credit risk of SMS enterprises in our country. In part three,we introduced the definition, characteristics and role of SMS enterprises in our country.The characteristics included three aspects: business characteristics, financialcharacteristics and credit characteristics. The role included promoting economicdevelopment, expanding employment, optimizing industrial structure and technicalinnovation four aspects. In pare four, we built a Loan credit risk evaluation model ofSMS enterprises and then tested it. In part five, the main content was the conclusionand deficiency of the paper. Then we gave some policy suggestions, mainly including:setting up perfect foundation database, the quantitative and qualitative methods,establishing a dynamic adjustment model, and establishing an independentmanagement department.