在线网络借贷投资决策模型及实证研究

发布时间：2018-06-26 00:34

  本文选题：网络借贷 + 投资决策　； 参考：《运筹与管理》2016年02期

【摘要】：P2P网络借贷作为电子商务在金融领域的延伸与应用,近年来得到广大学者的关注.但是目前的理论研究中,鲜有从投资者信息挖掘的角度进行投资决策分析.本文提出一个新颖的方法,即投资者构成分析方法,通过分析贷款的众多投资者信息遴选出最有价值的投资,辅助投资者进行投资决策.首先从投资者的历史投资收益率、风险偏好以及投资经验三个维度构建投资者档案(investor profile),进而基于投资者档案构建投资者构成分析模型,最后通过美国最大的在线网络借贷网站Prosper的数据,对本文提出的构想及模型进行了实证研究.实验结果表明本文提出的利用投资者构成分析的方法辅助投资者进行投资决策是可行的,文中构建的模型表现出良好的预测能力,能够有效地筛选出有价值的投资.
[Abstract]:As an extension and application of e-commerce in the field of finance, P2P network lending has been concerned by many scholars in recent years. However, in the current theoretical research, there are few investment decision analysis from the angle of investor information mining. In this paper, a novel method, investor composition analysis method, is proposed to select the most valuable investment by analyzing the investors' information of the loan, and to assist the investors to make investment decisions. Firstly, from the three dimensions of investors' historical investment return rate, risk preference and investment experience, the paper constructs the investor file (investor profile), and then builds the investor composition analysis model based on the investor file. Finally, through the data of Prosper, the largest online lending website in the United States, this paper makes an empirical study on the concept and model proposed in this paper. The experimental results show that it is feasible to use the method of investor composition analysis in this paper to assist investors to make investment decisions. The model constructed in this paper shows good predictive ability and can effectively screen out valuable investments.
【作者单位】： 大连理工大学管理与经济学部;
【基金】：国家自然科学基金资助项目(71402014) 教育部人文社科基金资助项目(14YJCZH044)
【分类号】：F713.36;F831.2
，

本文编号：2068236