我国担保企业个人贷款违约预警模型构建研究

Research on the Early Warning Model Construction of China's Guarantee Enterprise in Personal Loan Defaults

本文选取担保公司个人贷款业务数据为研究样本,建立个人贷款的违约预警模型,并检验违约预警模型的准确性。研究结果表明:年收入、年龄、共同还款人、贷款金额、还款期限、双B汽车等因素对个人贷款的违约会有显著影响;在收入越高且稳定、有共同还款人、有双B汽车、贷款金额小、还款期限长、年龄较大的前提下,贷款人越发不容易发生违约行为。本文研究成果以期对我国担保企业个人贷款风险控制有参考价值。

This paper selects the personal loans business data as the research samples. Firstly, we consider the selected data representation and effectiveness variables in the establishment of the sample. And then according to two groups of the normal and default, we divide the study sample into modeling group and test group. The former is used to establish the personal loan default prediction model, and the latter is used to test the accuracy of default warning model. Finally, the research results show that： annual income, age, common repayment, the loan amount, repayment period, double B cars and other factors will have a significant impact on personal loans default; under the premise of the lenders with the higher and stable income, common repayment, double B cars, less loan amount, long repayment period and older, they will have the less likelihood of default occurrence. The research has important reference value on the personal loans risk control of our guarantee enterprises.