Font Size: a A A

Research On Credit Risk Of P2P Lending Based On Neural Network Model

Posted on:2017-02-18Degree:MasterType:Thesis
Country:ChinaCandidate:L Y XuFull Text:PDF
GTID:2359330503972602Subject:Asset assessment
Abstract/Summary:PDF Full Text Request
Peer to peer lending refers to the transaction pattern that the individuals can borrow money from other people through the Internet platform. Since 2006, the domestic P2 P industry has developed rapidly. At the end of last year, the number of domestic P2 P industry's operating platforms is as high as 2595. The net loan volume of the year has achieved nine hundred and eighty billion. Behind the rapid development of the industry, is the continuous accumulation of systemic risk, among which the largest and most important risk category is credit risk. Compared with the ordinary financial lending companies such as commercial banks, securities firms, the P2 P industry's risk management level is lower. The reason mainly lies in that the existing P2 P industry is in a dilemma of insufficient development and talent shortage. The P2 P platforms generally lack professional credit evaluation personnel, and the borrower's credit level cannot be assessed accurately. In the absence of their own credit risk assessment system, the development of many platforms are limited.Taking the current situation of P2 P industry into account, this paper assesses the credit risk of different borrowers based on the operation model of net loan platform. According to the basic principle of BP neural network, the feasibility of evaluating the borrower's credit risk is studied. On this basis, this paper relying on both the basic characteristics of P2 P lending and traditional personal credit evaluation system, selects the appropriate indicators to build the P2 P borrowers' credit risk assessment system. This paper introduces in detail and designs steps of BP neural network evaluation model.In the empirical study, this paper takes the P2 P platform “renrendai” as the data source, training and testing the builded BP neural network credit evaluation model. The difference between the test results and the borrowers' true credit risk level is not large, indicating that the BP model can be used to assess the credit risk of P2 P lending.Finally, this paper puts forward the corresponding suggestions on the current situation of P2 P lending, and gives some suggestions and directions for the improvement and modification of the model.
Keywords/Search Tags:Peer to peer lending, Credit risk, BP neural network
PDF Full Text Request
Related items