Research On Credit Risk Evaluation Of SMEs Under The Supply Chain Finance Model

In recent years,the supply chain financial model has attracted the attention of the government,commercial banks,SMEs and other practical and academic circles as a way to alleviate the financing difficulties of SMEs.From the perspective of policy documents,the country is paying more and more attention to the implementation of supply chain work.In a sense,this model makes SMEs relatively transparent in financing,but at the same time,commercial banks also face greater risks when facing supply chain financing.Compared with traditional financing,the credit risk of SMEs under the supply chain finance model involves many influencing factors and spreads faster.At present,the main problems of credit risk assessment of supply chain financing SMEs are summarized into two aspects.First,in terms of evaluation index system,there is a lack of targeted index system,the evaluation index system is not perfect,and the weight of the evaluation index system is biased.Second,evaluation models are mostly subjective evaluation models,which rely on expert experience.Although experts have sufficient experience,the method is subjectively heavy and lacks objectivity.Firstly,this article compares the difference between the supply chain financial model and the traditional financial model,and finds that there are many factors affecting credit risk under the supply chain financial model,and the spread speed is fast,but it also has a certain degree of controllability.Secondly,use frequency analysis and correlation test to remove the indicators with strong correlation and establish a more objective and comprehensive index system,which mainly includes 5 first-level indicators: the status of financing enterprises,the status of core enterprises,the status of assets under financing,the status of supply chain,and the macroeconomics.After comparing various credit risk evaluation methods,it is found that the genetic algorithm-based BP neural network has the advantages of strong generalization ability,high fault tolerance,and broad assumptions,and can solve the standard BP neural network easy to fall the problem of local minima.Therefore,the BP neural network optimized based on genetic algorithm has established an evaluation model suitable for the credit risk of small and medium-sized enterprises under the supply chain financial model.Finally,the established model will be empirically and effectively analyzed in two aspects.First,according to the established index system,the five-year data of 28 auto companies on the New Third Board were selected for model calculation and analysis.The results show that the BP neural network optimized based on genetic algorithm is more accurate than the standard BP neural network in prediction set high.Second,a specific case analysis was conducted,and the results showed that the evaluation results of the BP neural network model based on genetic algorithm optimization were basically consistent with the actual situation.The research conclusion of this paper provides a certain reference for the evaluation of credit risk of small and medium-sized enterprises under the supply chain finance model,and helps commercial banks to better develop supply chain finance business.