基于Logistic回归模型的个人小额贷款信用风险评估及应用
Credit Risk Assessment of Personal Small Loan Based on Logistic Regression Model and Its Application

构建二分类Logistic信用风险评估模型，运用光大银行某分行样本数据，评估商业银行互联网金融个人小额贷款信用风险。结果显示：客户性别、学历、年龄、收入、职业、属地等因素对个人小额贷款信用风险影响显著。其中，年龄、收入、学历等与客户信用等级呈正向关系，女性信用风险显著低于男性，持有信用卡、存贷比越低的客户其信用等级越高；一、二线城市客户的履约率普遍高于县地级市客户的履约率。鉴此，商业银行应对互联网金融个人小额贷款信用风险进行有效规避和分散。
According to the actual sample data collected from one branch of Everbright Bank of China, this paper built a two-classification Logistics credit risk assessment model on personal small loan credit risk assessment. Empirical evidence showed that: age, gender, income, occupation, educational background, credit card holding, the LDR customer are factors that very significantly affect personal small loan credit risk with age, income stability, and education level positively related with the risk; women's credit risk is significantly lower than that of men's, especially for those holding a credit card, or those with lower the LDR women customers; the performance rate of first-tier and second-tier cities is generally higher than that of county level city clients. The bank should take specific measures to effectively avoid and diversify risks.