摘要
针对P2P网络信贷平台信用风险特点,以借款人违约概率为被解释变量,运用Logistic回归方法建立新进借款人信用风险评估模型。分析认为,年龄、性别、岗位职位、收入、借款用途、工作认证和实地认证指标应作为评价个人信用风险的主要依据。基于回归结果认为,扩展信用评估指标类型、加大线下审核力度是对平台新进借款人信用风险防控的有效策略。模型的构建为网络信贷行业建立统一的借款人信用评估体系提供了合理依据。
According to the characteristics of P2 Plending credit risk,this paper establishes the new borrower credit risk assessment model by means of Logistic regression method with the borrowers′default probability as the explanatory variable.It is concluded that age,gender,job position,income,purpose of loaning,work certification and field authentication are the main factors causing the borrower credit risk.This paper finds that extending credit evaluation index types and intensifying offline review are the effective strategies for preventing credit risks based on the regression results.The construction of model can provide reasonable grounds for designing and constructing unified credit evaluation system for borrowers in P2 Plending industry.
出处
《江苏科技大学学报（社会科学版）》
2016年第3期102-108,共7页
Journal of Jiangsu University of Science and Technology(Social Science Edition)
