商业研究

• 商经理论 • 上一篇    下一篇

P2P网络借贷借款人违约风险影响因素研究

李杰1, 刘露1, Chao-Hsien Chu2   

  1. (1.河北工业大学 经济管理学院,天津 300401; 2.美国宾夕法尼亚州立大学 信息科学与技术学院,美国 宾夕法尼亚州 16802)
  • 收稿日期:2018-03-16 出版日期:2018-09-10
  • 作者简介:李杰(1973-),女,河北河间人,河北工业大学经济管理学院教授,工学博士,研究方向:互联网金融、商务大数据分析;刘露(1992-),女,河北沧州人,河北工业大学经济管理学院研究生,研究方向:互联网金融欺诈识别、商务大数据分析;Chao-Hsien Chu(1951-),男,美国华裔人,美国宾夕法尼亚州立大学信息科学与技术学院教授,管理学博士,研究方向:智能信息处理。
  • 基金资助:
    国家社会科学基金资助项目,项目编号:16FGL014。

Influence Factors of Borrower Default Risk in Peer-to-Peer Lending

LI Jie1, LIU Lu1, Chao-Hsien CHU2   

  1. (1. School of Economics and Management, Hebei University of Technology, Tianjin 300401, China; 2. College of Information Sciences and Technology, The Pennsylvania State University, PA, 16802, USA)
  • Received:2018-03-16 Online:2018-09-10

摘要: 借助于互联网信息技术的发展,P2P网络借贷推动了互联网金融以及普惠金融的发展,有关平台、出借人如何做出准确风险评估、防范借款人违约风险变得尤为重要。本文以融360平台提供的4738名借款人借贷数据为研究样本,运用Logistic回归模型分析借款人违约风险的关键因素,就违约借款人的具体特征以及其影响因素展开分析。研究结果表明:在还款能力方面,经济特征中借款人总收入、总支出、工资收入因素对借款人是否发生违约有显著影响;在还款意愿方面,性别、借款额度、借款金额以及拖欠金额对借款人违约风险产生显著影响;在线上浏览行为方面,借款人最少浏览网页数量和网站访问次数是P2P网络借贷借款人违约风险显著因素。因此,风险监管部门应建立关键信息共享机制,明确审查范围,落实审查重点,通过建立违约风险评估模型降低平台和出借人的经济损失,推动P2P网络借贷行业的健康发展。

关键词: P2P网络借贷, 违约风险, Logistic回归模型

Abstract: With the development of Internet information technology, P2P lending has promoted the development of Internet finance and inclusive finance. It is particularly important for the relevant platforms and lenders to make an accurate risk assessment and prevent the default risk of borrowers. Taking 4738 borrowers′ lending data provided by Rong 360 platform as research samples, the paper adopts Logistic regression model to analyze the key factors of borrower default risk, and discusses the specific characteristics of the delinquent borrower and its influence factors. The empirical results show that the borrower′s total income, total expenditure and wage income among economy features have significant impacts on borrower′s default risk in terms of repayment ability; borrower′s gender, loan limit, loan amount and delinquent amount have obvious effects on borrower′s default risk in term of repayment willingness; the number of the fewer web page and website visits are important factors of borrower′s default risk in terms of online browsing behavior. Therefore, the risk supervision department should establish a key information sharing mechanism, clarify the review scope, implement the key points of examination, and then reduce the economic losses of platform and lenders by constructing a default risk assessment model to promote the healthy development of the P2P lending industry.

Key words: P2P Lending, default risk, Logistic Regression Model