基于Logit模型的P2P网络借贷平台借款人信用风险影响因素研究

发布时间：2018-01-14 19:21

本文关键词：基于Logit模型的P2P网络借贷平台借款人信用风险影响因素研究　出处：《哈尔滨商业大学》2017年硕士论文　论文类型：学位论文

更多相关文章： P2P 网络借贷 借款人信用风险 Logit

【摘要】：P2P(peer-to-peer)网络借贷,是一种借助电子商务的专业网络平台,是个人与个人之间互为借贷的小额借贷交易。P2P网络借贷由于具有门槛低、主要从事无抵押借贷、借款方式相对透明等特点,备受中小微企业和个人的追捧。特别是在美国2008年金融危机之后的几年内,传统金融机构融资低迷,但是,网络借贷的发展浪潮持续增高。伴随着互联网金融以及互联网金融产品的高速发展,P2P网络借贷违约率即信用风险成为大家人以及社会所关注的焦点,高信用风险成为网络借贷自身发展的最大瓶颈。本文采用排序选择模型,基于excel VBA数据挖掘技术,编写宏程序,通过网页固定抓取数据,分别从国内最早的P2P网络借贷平台——拍拍贷网站,及目前发展最好的网站——人人贷网站上截取贷款数据,选取了借款人个人特征信息(年龄、性别、借款人职业)、借款人交易特征信息(历史借款记录、借款目的)、平台评价信息(信用等级、贷款规模、利率、贷款期限、每月还款额)和借款人投标信息(中标次数、流标次数)四个方面作为信息数据,并运用Logit模型对P2P网络借贷借款人的信用风险影响因素进行了实证分析。研究结果显示:(1)借款人职业与借款人信用风险存在着显著的正相关关系。借款人职业越稳定,其信用风险越大。(2)借款人借款记录与借款人信用风险存在着显著的正相关关系。借款人目的与借款人信用风险之间存在着显著的负相关关系。借款人借款的信息越真实,借款动机可靠性越强,借款人信用风险越小。(3)借款人信用评级与借款人信用风险存在显著的负相关关系。借款人贷款规模与借款人信用风险负相关,借款人利率和每月还款额与借款人信用风险正相关。(4)借款人中标次数与借款人信用风险存在着显著的正相关关系;借款人流标次数与借款人信用风险存在着显著的正相关关系。借款人在平台中越活跃,其信用风险就越大。本研究结果,既可以为防范P2P网络平台信用风险提供新的思路,也可以为完善我国P2P网贷行业治理提供新的经验证据。
[Abstract]:P2Ppeer-to-peer) online lending is a professional network platform with the aid of electronic commerce. Peer-to-Peer network lending is a small loan transaction between individuals and individuals. Because of its low threshold, mainly engaged in unsecured lending, borrowing methods are relatively transparent and so on. Especially in the years following the 2008 financial crisis in the United States, the financing of traditional financial institutions was depressed, but. With the rapid development of Internet finance and Internet financial products, P2P network loan default rate, that is, credit risk, has become the focus of people and society. High credit risk has become the biggest bottleneck in the development of network lending. This paper adopts the sorting selection model, based on excel VBA data mining technology, compiles macro programs, and grabs data through web pages. From the earliest domestic P2P network lending platform-PPDAI website, and the best developed website-peer-to-peer lending website to intercept loan data, selected the borrower's personal characteristics information (age, gender). Borrower occupation, borrower transaction information (historical loan record, loan purpose, platform evaluation information (credit rating, loan size, interest rate, loan maturity). The monthly repayment amount) and the information of the borrower's bid (the number of winning bids, the number of the flow mark) are taken as the information data. Logit model is used to analyze the influencing factors of credit risk of P2P network loan borrowers. The results show that: 1). There is a significant positive correlation between the borrower's occupation and the borrower's credit risk, and the more stable the borrower's occupation. The greater the credit risk, the greater the credit risk.). There is a significant positive correlation between the borrower's loan record and the borrower's credit risk. There is a significant negative correlation between the borrower's purpose and the borrower's credit risk. The stronger the reliability of borrowing motivation, the smaller the borrower's credit risk.) there is a significant negative correlation between the borrower's credit rating and the borrower's credit risk, and the scale of the borrower's loan is negatively correlated with the borrower's credit risk. The borrower's interest rate and monthly repayment amount are positively correlated with the borrower's credit risk. (4) there is a significant positive correlation between the number of times the borrower wins the bid and the borrower's credit risk. There is a significant positive correlation between the number of logovers and the credit risk of the borrower. The more active the borrower in the platform, the greater the credit risk. It can not only provide new ideas for preventing the credit risk of P2P network platform, but also provide new empirical evidence for perfecting the governance of P2P network loan industry in China.
【学位授予单位】：哈尔滨商业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F724.6;F832.4

【参考文献】