Research On Credit Default Risk Of Credit Borrowers In P2P Network In China

P2P Internet credit refers to personal and personal credit.It refers to investors who have idle funds and financial ideas.They borrow a third-party platform to act as an intermediary and lend to other borrowers with borrowing needs.As of February 2018,the platform The cumulative number has reached 6054,and the industry's turnover has reached 6611.144 billion yuan.P2 P network credit has become an important channel for individuals and small business owners to carry out financing,and it is a kind of emerging financial model in China in recent years.The borrower of the demand provided a new platform,but as a new financial model,the P2 P platform also has many problems in its development.The risk management capability is also greatly insufficient,and it also brings great risks to the market while it develops.In particular,there are credit risks on the platform.So far,many lending platforms have not yet established a good credit default risk assessment model.Studying borrowers' credit default risk has certain practical significance.This article first briefly introduced the basic concept of P2 P network credit and the current status of P2 P network credit analysis.Secondly,based on the statistical analysis of state-owned commercial banks and tapping loan credit default risk indicators,a statistical analysis of the data was conducted.In the final indicator system of this paper,the number of repayments is added to the index system.After adding this variable,the model has a higher fitting effect and the model is more accurate.Finally,in several common models for assessing credit default risk,Compare the advantages and disadvantages to determine the model used in this article.In the construction of the logistic regression model,descriptive statistical analysis of the data was performed first.Then the logistic regression model was used to filter backward indicators by the Wald method.Finally,the borrower's 8 indicators were determined as final variables,among which the borrower's repayment.The number of pens is a new variable.During the model construction process,it was found that the credit rating,gender,annual interest rate,number of repayments,and overdue times are important variables affecting the borrower's default,and when predicting the credit default risk of the borrower,If the predicted probability is greater than 0.5,the borrower is predicted to be in default,and if the predicted probability is less than 0.5,the borrower is predicted not to be in default.Finally,test the effect of the model.Finally,in the borrower's various indicators,the credit rating,overdue number,borrowing interest rate,number of repayments,and gender have a greater impact on whether the borrower defaulted,and the borrower's academic qualifications,marital status,The nature of the work has little effect on whether the borrower defaults.Therefore,when assessing the credit default risk of the borrower,the platform should focus on the borrower's credit status.