Research On Risk Early Warning Of P2P Network Lending Platform In China

P2P network lending is a typical representative of Internet finance.Its development has a huge impact on the Internet finance industry.At the same time,P2P network lending has always been faced with a series of risk issues in the process of rapid development in China,which caused some frequent occurrences of P2P network lending platforms,such as breaking out of making off with money,closing,difficulty in making withdrawals.At present,the laws and regulations of the entire P2P network lending industry have been introduced,but before the specific implementation and further improvement,there are still some insecurity factors in the P2P network lending industry.Therefore,how to carry on risk early warning of P2P network lending platforms to protect the interests of investors effectively is still very necessary.This paper first introduced the development stage of P2P network lending industry in China,analyzed the main risks faced by P2P network lending platforms,introduced the practice in risk early warning of Internet financial in China,the purpose is to draw forth the importance of risk early warning to P2P network lending platforms.Secondly,this paper built a P2P network lending platforms risk early warning system in China.Combined with the comprehensiveness,representativeness,and availability of the indicators,this paper analyzed the main risks of the P2P network lending platforms and constructed a P2P network lending platforms risk early warning index system,the index system includes five first-level indicators and 30 second-level indicators.Then this paper built risk early warning models of P2P network lending platforms in China based on Logistic regression,BP neural network and SVM respectively.To adapt to appropriate supervision and classification regulation for the supervision department,the standard for dividing the risk early warning interval of each model was proposed.Again,the paper used the risk warning models for empirical analysis.The paper used the principal component analysis method to comprehensively evaluate the P2P network lending platforms,so as to determine the sample set for the risk early warning system of P2P network lending platforms based on Logistic regression,BP neural network and SVM for empirical analysis.The research results show that,combined with the P2P network lending platforms risk warning index system,the result of the risk warning system who uses BP neural network or SVM of P2P network lending platforms is effective,and the police judgment is basically consistent with the actual situation.Therefore,we can use the risk early warning system billt in this paper,which based on BP neural network or SVM,to make risk early warning for P2P network lending platforms.Finally,in order to prevent the risk of P2P network lending platforms effectively,this paper put forward the countermeasures for improving the risk early monitoring and warning of P2P network lending industry,including improving the data quality and optimizing the early warning methods.What's more,this paper put forward the suggestions for reducing the risk of P2P network lending industry on two levels:the government and the platform.