摘要
P2P platform default risk seriously affects the returns of investors, which may cause systemic financial risks. The existing literature mostly focuses on borrower risk, ignoring the research on P2P platform default risk. This paper uses signal theory and data mining-related methods to study the default risk prediction of P2P platforms that integrate soft and hard information signals in different economic environments. First, using the cluster analysis method, the macroeconomic environment of P2P platforms is studied. Second, from the perspective of signal costs, signal theory is used to analyze the impacts of soft and hard information risk signals on platform default in different economic environments. Finally, by integrating the lasso and stacking methods, a LAS-STACK model is proposed to study the prediction of P2P platform default risk in the high-dimensional unbalanced data context. The conclusions of this paper show that the fusion of soft and hard information can better predict the default risk of P2P platforms, especially during periods with low economic levels. Additionally, the LAS-STACK model has a better prediction ability for the P2P platform default risk in the high-dimensional unbalanced data context. This study can improve the ability of regulators and P2P platforms to warn and manage default risks in a specific economic environment and protect investors' returns.
学科与关键词
文献加载失败,请重试
重试