The Prepayment Of P2P Lending And Research On Its Investment Return

In the P2P lending market, based on the interest rate and the probability of payment, investors do their diversified and decentralized portfolios by the credit risk assessment of the borrower. Credit risk assessment concentrates on whether the borrower will pay the loan and interests, which helps investors to reduce their overall risk, but ignores the actual return on investment ultimately. Prepayment behaviors in P2P lending will cause the borrower to repay the loan that the actual cash flow is faster than the expected ones, which results in the uncertainty of cash flow to ther investors, affects the use of cash effectiveness and the ultimate return on investment.Current researches on P2P lending focus on personal credit assessment and loan default, but are lack of prepayment behaviours reasech on the borrowers without default, prepayment and default are two of the main causes affecting the profitability of a loan. It’s not advsible for the lenders’to formulate investment decisions which only depends on the borrower’s willingness to repay. The investment return is not in accordance with the expected return. So there is a need to make deeper classification of borrowers by requiring prepayment risk and to estimate the actual repayment behaviors. Thus we can calculate the actual investment rate of return in the P2P market.In this paper, we propose a feasible method to study the prepayment behavior. There are two research models, one is P2P classification model and ther other is P2P pricing model. We do our research by the following steps:first, classify the P2P prepayment behavior by decision tree algorithm in data mining, distinguish borrowers with their repayment, prepayment and default behaviours, and summarize their characteristics in prepayment behavior; second make predictions of the repayment period through M5P algorithm; third calculate cash flow of each period in the prepayment process; fourth use discounted cash flow method to finally obtain the actual investment yield in P2P lending. The first step of our study is to build a prepayment classification model for borrowers which provides an assessment of reference for judging borrowers who will prepay. Based on the classification model, next steps are to build prepayment pricing model, thus to estimate the quantitative results, which is the final return on investment.Research contributions of this paper are as follows:First, the prepayment behavior occurres in the mortgage securitization market, by observing the actual behavior and loan repayment status of the P2P lending market, we study the prepayment behavior in it to mine borrower’s willingness to repay. Second, we propose a feasibility method to make P2P prepayment behavior classification and asset pricing, generate a new understanding of P2P lengding which is similar to fixed-income investment assets. Third, as the securities issued in the primary market and the secondary market for sale, the occurrence of such P2P prepayment makes its value changed, this paper proposes a new pricing model to estimate actual rate of return on investment, which provides a reference for asset securitization pricing in P2P lending market. Fourth, the establishment prepayment risk assessment will help improve the prediction accuracy of returns for the investors in the P2P lending market, which can helo investors to balance risk and to achieve expected benefits, and to establish strategy combinations efficiently.