Analysis Of Influencing Factors On Default Rate Of P2P Borrower

P2P online lending is a form of lending that is based on Internet thinking and independent of the formal financial system.At the end of December 2016 the number of China's P2P online business lending platform has reached 3537,the cumulative turnover has reached 2 trillion and 804 billion 938 million yuan.The rapid growth of the number and the lack of supervision have increased the risk of the P2P platform.At present,Chinese economic structure is undergoing in-depth adjustment,economic growth rate declined,P2P net loan industry credit risk exposure continues to rise,the lenders' default rate continued to rise.P2P net loan platform profit by operating risks,therefore,risk management technology is the core competitiveness of P2P net lending platform,and is also the main thread of the development of P2P net loan industry in the future for a period of time.To study how different reasons influence the default rate,first thing was to know the causes of credit risk arising from P2P network loans,and analyzed the specific factors that affect the default rate of borrowers by analyzing the borrower's information,combining the actual situation of P2P net loans.Based on the domestic and foreign research results,the initial selection of the borrower qualifications,age,loan amount,loan rate of 10 indicators were analyzed,and through principal component analysis,the number of variables was reduced to 4.Specific expression was got from the establishment of Logistic model,which showed that the height,marital status,age and condition,certification rating is inversely proportional to the default rate,lenders who were in higher education,older,married men,the default rate will be lower.And the loan amount,loan period,income and the loan interest rate is proportional to the risk of default of the borrower.The higher interest rate and longer duration of the lending would lead to higher risk.This was consistent with our previous hypothesis,and also conformed to the actual situation.At the same time,the sample data and the out of sample data group also verified the higher prediction accuracy of the model.