Rural Credit Cooperatives Credit Risk Evaluation System Based On Logistic Regression Model

Recently the credit risk measurement in Rural Credit Cooperatives is the fivegrades classification, which is more mature and has been put into use. Theimplementation of the five grades classification effectively promotes the credit assetrisk management of the rural credit cooperatives. But the qualitative method is easilyaffected by subjective evaluation of human factors. It can not accurately define riskdivision scope. The personal credit risk evaluation system of rural credit cooperativesis based on the five grades classification. And it is introduced the Logistic regressionanalysis, which is based on data mining technology. In personal credit risk evaluationsystem, data acquisition, data selection, data cleaning, data transformation, dataprocessing, data sampling, Logistic regression model testing, model evaluation arespecifically analyzed; On this basis, the400cases of rural credit cooperatives areimported, which include200good sample and200bad sample. Then the miningprocess of sample data has been carried out. And the Logistic regression model isestablished. The application, testing and evaluation of the personal credit riskevaluation system are achieved. In general, the personal credit risk evaluationsystem has upgraded the evaluation function of the five grades classification.Thefunction of personal credit risk quantitative assessment can solve the problem of thetime lag in the five grades classification management. So the personal credit riskevaluation system can provide guidance role for the management of credit risk, whichhas a strong practical significance.