Data-driven Investment Decisions in P2P Lending: Strategies of Integrating Credit Scoring and Profit Scoring
Date of Award
Doctor of Philosophy in Analytic and Data Science
Statistics and Analytical Sciences
In this dissertation, we develop and discuss several loan evaluation methods to guide the investment decisions for peer-to-peer (P2P) lending. In evaluating loans, credit scoring and profit scoring are the two widely utilized approaches. Credit scoring aims at minimizing the risk while profit scoring aims at maximizing the profit. This dissertation addresses the strengths and weaknesses of each scoring method by integrating them in various ways in order to provide the optimal investment suggestions for different investors. Before developing the methods for loan evaluation at the individual level, we applied the state-of-the-art method called the Long Short Term Memory (LSTM) model to predict the default risk of P2P lending at the aggregated level, thus providing investors a thorough understanding of the status of the whole P2P market first. Then we proposed three methods based on the integration of credit scoring and profit scoring in order to sort out the top loans. The first method is the two-stage evaluation system, focusing on integrating the credit information into profit scoring. The second method is the profit-sensitive learning method, focusing on integrating the profit information into credit scoring. The third method is the bivariate model, aiming at simultaneously evaluating the risk and the profit by taking into account their correlation. Experimental studies show that the proposed three methods demonstrate their superiority over the traditionally utilized credit scoring and profit scoring techniques in terms of identifying the loans with a higher profit without introducing extra risk.
Finance and Financial Management Commons, Longitudinal Data Analysis and Time Series Commons, Multivariate Analysis Commons, Risk Analysis Commons, Statistical Methodology Commons, Statistical Models Commons, Theory and Algorithms Commons