P2P Online Lending Risk Early Warning System Study

In 2007, “Paipai Dai”,the first P2 P online lending platform of our country, set up in Shanghai. Then, “Yi Xing” and “Hong Ling Capital” were all founded. P2 P online lending attracted more and more people with a low threshold and flexibility, ect. And it developed on a large scale in a short time. However, on the rapid development of online lending industry, it also left a huge immature risk due to the lack of supervision and risk control. Currently, the weak risk control ability is the majority problem of P2 P online lending platform. Therefore, P2 P online lending must improve its risk control capability and establish an effective risk control early warning mechanism. This article build a P2 P online lending risk early warning theoretical framework based on the analysis of the risk of P2 P online lending platform. And make empirical analysis on Yiren Dai.com. Finally, provide advice to its development.This paper mainly has five parts.In the first part, mainly introduce the concept, patterns, development history and current situation of P2 P online lending.In the second part, analyze the risk of P2 P online lending from the government level, market level and internal level. Government level includes regulatory risk and legal risk. Market level includes credit risk and liquidity risk. Internal level includes operation risk and information system risk. Then, summarize the main risk control methods of P2 P online lending platform based on all level risk, which mainly include user credit review, diversity investment, overdue accounts collection system, risk reserve system, guarantees and mortgage system.In the third part, mainly build the theoretical framework of risk early warning system of P2 P online lending. First, establish the risk early warning indicator system of P2 P online lending in terms of the development index rating of net credit home. This paper initially selects 20 indicators. Then, use the factor analysis to evaluate the risk, so as to determine the risk evaluation standard. Finally, choice BP neural network model by reference the risk early warning of commercial banks. And introduce the feasibility of BP neural network model and its algorithms and procedures.In the fourth part, make empirical analysis on Yiren Dai.com. First, use cluster analysis to filter indicators,and form a concise risk early warning indicators system. Then, make factor analysis to these indicators, and arrived three public factors, which are loan balance factor, dispersion factor and narrow liquidity factor. At the same time, divide the months into steady and risk according to the average of the principal component comprehensive score for each month. Finally, use the MLP method to build and test BP neural network model, and then use the model to warn risk.In the fifth part, mainly provide advice of P2 P online lending risk management. First, establish risk early warning management system. Then, promote risk early warning regulation. Finally, establish the credit reporting system gradually.