Research On Credit Risk Quantitative Rating System Of Auto Loan Services In Zhejiang Mingfa Company

With the successive introduction of a series of auto finance policies to promote the growth of auto consumption in recent years,more and more customers choose to purchase cars by means of Consumption staging,and the auto consumer finance market has entered a booming stage.At the same time,automobile consumer financial institutions are increasing,and competition is becoming increasingly fierce.To a certain extent,financial institutions have relaxed their requirements for customer qualification review,which has resulted in various risks such as operational risk,credit risk and market risk.The non-performing rate of the loan business also increased,causing serious losses to financial institutions.The non-performing rate of the loan business also increased,causing serious losses.The non-performing rate of auto loan business of financial institutions has also risen,causing serious losses.In this case,the risk management of related financial institutions is facing new challenges,and the accuracy and standardization of risk measurement has become one of the more important management strategies.Based on this background,the purpose of this thesis is to study the quantitative credit risk rating of Zhejiang Minfa Company's car loan business,and to provide practical guidance and quantitative basis for solving problems such as low accuracy,low efficiency and poor stability of customer audits.At present,Zhejiang Minfa Company's credit risk rating for car loan customers is still mainly based on human experience and subjective judgments.It has problems such as poorly targeted credit risk management,unclear key risk indicators,and no quantitative standards.Based on this,the paper first uses the chi-square test,independent T-test,etc.to analyze and arrange the historical data of its car loan business,and then uses the decision tree model to recursively select the optimal feature principle to bin the indicators.Then use Logistic regression model to operate on 37 indicators of 23923 customers,screen out 15 key risk indicators that affect credit risk,build a quantitative credit risk rating index system,and formulate quantitative rating indicators based on the score distribution and corporate risk principles System scheme.Finally,an additional 100 customer samples were collected to verify the effectiveness of the credit risk rating system.The verification results show that the accuracy rate of customers' overdue recognition in the quantitative rating system reached 93%.The rating system constructed in this paper can accurately and effectively identify high-risk customers,and can provide a scientific quantitative basis for customer credit risk management.The main conclusions and values of the research can be summarized into three aspects.The first,the quantitative rating system not only helps to improve the accuracy of the customer's credit risk review,but also can prevent the risk loss caused by the ability of auditors or ethical factors;the second,the quantitative rating system can The reduction of more than half has significantly reduced the company's labor costs.The third,the quantitative rating system provide guidance for the company's scientific management of customer risks,to implement a credit risk management system based on quantitative ratings and supplemented by manual verification to enhance the credit risk review process.Stability,ensuring that risk management is less subject to human intervention,and helping the company expand its market,enhance the company's business competitiveness,and increase the company's market share of car loan products.