摘要: |
从P2P平台信用风险角度出发,以借款人风险控制为研究目标,构建借款人信用评价指标体系,并利用美国P2P网络借贷平台Prosper上的数据建立基于 Logistic 回归的借款人信用风险评价模型。实证分析表明:是否有房产、贷款创立时长、借款利率对借款人信用风险有着比较大的影响,而借款金额、信用评级、借款期限、借款用途对借款人信用风险没有特别明显的影响。 |
关键词: P2P网络借贷 信用风险 互联网金融 Logistic模型 |
DOI:10.13216/j.cnki.upcjess.2017.06.0003 |
分类号:F832.4 |
基金项目:福建省中青年教师教育科研项目(JAS161010) |
|
Research on Credit Risk Assessment of Online Lending Borrower Based on Logistic Regression Model |
XU Huiting
|
(Business Department, Xiamen City University,Xiamen, Fujian 361000, China)
|
Abstract: |
From the perspective of the credit risk of the P2P platform and to study the borrower 's risk control, this paper proposes a set of borrower credit evaluation index system . Then an evaluation model is built based on logistic regression, using the data of Prosper. The empirical results show that IsBorrowerHomeowner, AgeInMonths and BorrowerRate have greater impacts on the borrower 's credit risk, while AmountBorrowed, ProsperRating, Term and Category have no significant effect on the borrower 's credit risk. |
Key words: online P2P lending credit risk internet finance Logistic Regression model |