Research On Credit Risk Identification Of P2P Net Loan Borrowers In China

Since the introduction of P2 P network lending in China in 2007,P2 P internet lending has developed rapidly in the short term.P2 P network lending is driven by the rapid development of the Internet economy and the number of operating platforms presents a blowout growth.Between 2007 and 2015,the P2 P network lending operating platform from the beginning of the first to the present 2595.And 2015 single year increased by 1020,the absolute increase is far more than any time in the past.According to statistics,the platform average monthly turnover of 49.26 billion yuan,and its number has been in the state of expansion.However,due to the P2 P network borrowing platform has a low entry threshold,the lack of industry standards,strict supervision and so on,which makes the P2 P network lending platform serious information asymmetry problem,so the platform will often appear the "Paolu" phenomenon.According to statistics,as of June 2015,the domestic credit problems of the P2 P network lending platform more than 550.At present,foreign financial markets to the P2 P network lending to provide a healthy development environment,in addition,a comprehensive regulatory mechanism also contributed to the development of foreign P2 P network lending.But the development of P2 P network lending companies by the actual certain restrictions.As a result of the emerging industry,its risk control ability can't be compared with the traditional banks.In general,there are two main risks to P2 P network lending: basic and specific risks.Basic risks are legal risk,credit risk and regulatory risk.Specific risks are investment risk,information asymmetry risk,and so on.One of the most serious is the credit risk.Many P2 P network lending platforms still follow the credit identification model of traditional commercial banks,and do not develop a credit risk identification model which is more suitable for P2 P network lending according to the characteristics of P2 P network lending.Based on the data of "Paipaidai",combined with the existing credit risk identification methods at home and abroad,the index system of credit risk identification is established.On the basis of SPSS and MATLAB,lots of Logistic r-egression methods were used to establish the A-F regression equation with the order of the mirror grade "E" as the reference level.And the significant influencing factors of each grade were obtained.Then the neural network model is established.First of all,the variables are normalized.Finally,the two methods are compared and summarized.According to this study,the above methods can be effective identification of lending credit risk.According to the calculate-on results,the credit risk level of the borrower can be effectively identified.And the ability of the two models to identify the borrower's credit risk is verified.In the follow-up study,suggestions were made on how to improve the credit risk of borrowers.In order to strengthen the risk recognition ability of the P2 P network lending,it should enrich the multi-level information authentication index and strengthen the effective supervision of the P2 P network lending.