Font Size: a A A

Credit Risk Assessment Of P2P Net Loan Borrowers

Posted on:2021-01-21Degree:MasterType:Thesis
Country:ChinaCandidate:Q XuFull Text:PDF
GTID:2439330611487316Subject:Probability theory and mathematical statistics
Abstract/Summary:PDF Full Text Request
P2P network loan is a new financial model which combines Internet with private loan.Its characteristic is that the borrower and the lender take the network credit company as the medium,in the process of the loan,the materials,contracts and procedures are all realized through the Internet.In recent years,P2 P network lending has developed rapidly in China,because it can not only meet the needs of personal funds,but also improve the utilization rate of social idle funds.P2 P network lending is a business that relies heavily on credit system and data analysis and processing.The P2 P network lending platform in western countries is relatively connected with credit system,and the credit data of borrowers is relatively accurate and comprehensive,so it can directly decide whether to lend through online.As the credit system of our country is not perfect and does not connect with the network lending platform,the bad debt rate of P2 P network lending platform is still high.As a result,many platforms have closed down or run away,leaving the money of the lenders dead,causing a very bad social impact.Therefore,it is an urgent problem to find a personal credit risk assessment method suitable for online lending platform.This paper first expounds the reasons and significance of the research on the credit risk of the borrower,and then introduces the P2 P network lending,including its development history and the causes of the personal credit risk of the borrower.Then,based on the evaluation index system of common online lending platforms,a personal credit risk evaluation system is established,which includes 14 indexes,including loan cycle,education background,monthly average income,company size,unit type,real estate situation,car property situation,marital status,position,working time,age,number of loan applications,number of successful loans and number of loan repayments.Then,using the decision tree support vector machine combination model,the credit rating of 599 loan records collected on Renren loan website is divided,which includes A-level,d-level,e-level and HR level.The result of an example shows that the accuracy of the model is high,and it can be used as a reference for the credit classification of P2 P network loan platform.Finally,21 samples with higher credit rating are selected from 599 borrowers,and entropy weight TOPSIS method is used to rank credit risk.Firstly,entropy weight method is used to calculate the weight of each evaluation index.This method can effectively reduce the subjectivity of index evaluation.Then TOPSIS method is used to rank the credit risk of borrowers,which can solve the problem of loan issuance sequence in the case of limited number of loans,and then help to reduce the risk of P2 P network loan platform.At the end of this paper,combined with the current situation of P2 P network loan platform,the future research direction on the healthy development of P2 P network loan platform is proposed.
Keywords/Search Tags:P2P network loan, Credit risk, Support Vector Machines, TOPSIS method
PDF Full Text Request
Related items