刘盼,李华康,孙国梓.基于短时多源回归算法的P2P平台风险观测方法[J].计算机科学,2018,45(5):97-101
基于短时多源回归算法的P2P平台风险观测方法
Risk Observing Method Based on Short-time Multi-source Regression Algorithm on P2P Platform
投稿时间:2017-04-17  修订日期:2017-07-09
DOI:10.11896/j.issn.1002-137X.2018.05.017
中文关键词:  P2P网络借贷,运营风险,时间窗,短时多源回归算法
英文关键词:P2P lending,Operation risk,Time window,Short-time multi-source regression
基金项目:本文受国家自然科学基金青年项目(61502247),公安部重点实验室开放课题(2015DSJSYS001),江苏省高校自然科学研究面上项目(14KJB520028)资助
作者单位E-mail
刘盼 南京邮电大学计算机学院软件学院 南京210003  
李华康 南京邮电大学计算机学院软件学院 南京210003  
孙国梓 南京邮电大学计算机学院软件学院 南京210003 sun@njupt.edu.cn 
摘要点击次数: 264
全文下载次数: 175
中文摘要:
      P2P网络借贷作为当代互联网金融领域中流行的借贷方式,具有借款金额小、还款周期长短不一的特点,导致传统的年度风险评估方法因时间粒度过粗而容易给 平台投资者造成损失。基于此,提出一种基于短时多源回归算法的网络借贷平台运营风险的动态评估方法。通过动态时间窗对借贷记录进行切分,并以线性回归来量化平台的动态风险指数。实验结果表明,该方法能够及时反映P2P平台的风险宏观运营情况,并向投资者提供平台的动态风险评估和预测指标。
英文摘要:
      Peer-to-Peer(P2P) lending is a popular lending way in the field of contemporary Internet finance.There are small lending amount and different repayment cyclelengths,which may easily cause the loss of platform investors due to annual risk assessment.This paper proposed a method to dynamically evaluate the operation risk of P2P platforms based on short-time multi-source regression algorithm.In this algorithm,dynamic time windows are used to split up the len-ding records and linear regression method is used to quantify the dynamic risk index of P2P platforms.The experimental results show the method can reflect the visible operation situation of platforms,and can provide dynamic risk assessment and forecast indicators of the platforms to investors.
查看全文  查看/发表评论  下载PDF阅读器