Asset mispricing in peer-to-peer loan secondary markets
Asset mispricing in peer-to-peer loan secondary markets
复制标题
DOI：
10.1016/j.jcorpfin.2020.101769
复制
发表时间：
2020-12
影响因子：
6.1
通讯作者：
M. Çaglayan;Tho Pham;Oleksandr Talavera;Xiong Xiong-Xiong
中科院分区：
文献类型：
--
作者：
M. Çaglayan;Tho Pham;Oleksandr Talavera;Xiong Xiong-Xiong
本研究调查了欧洲领先的 P2P 借贷平台 Bondora 在 2016 年至 2019 年期间是否存在定价错误的情况。通过实施机器学习方法，我们衡量 Bondora 二级市场上贷款转售成功的可能性，并将我们的预测与事后市场结果进行比较。观察到的差异揭示了两个现象，这与市场参与者对资产价格和相关基本面的不同看法有关：一些不可出售的资产被出售，而高可出售资产的转售并不成功。一旦卖家观察到买家的行为揭示了买家对资产价值的信念，他们的定价行为就会发生变化。我们的结果对于各种统计和机器学习方法都是稳健的。
This study examines the presence of mispricing on Bondora, a leading European peer-to-peer lending platform, over the 2016–2019 period. By implementing machine learning methods, we measure the likelihood of success for loan resale on Bondora's secondary market and compare our predictions with the ex-post market outcomes. The differences observed uncover two phenomena which are related to the diverging perceptions of market participants on asset prices and associated fundamentals: some non-saleable assets are sold, while the resale of highly saleable assets is not successful. Sellers' pricing behaviour changes once they observe buyers' actions revealing the buyers' beliefs about the value of the asset. Our results are robust to various statistical and machine learning methods.