Research On Credit Risk Assessment Model Of SMEs Based On Supply Chain Finance Model

With the advancement of economic system reform,the scale of small and medium-sized enterprises in China is growing day by day.As the most growth group in China ’s economy,small and medium-sized enterprises have a profound impact on China ’s economy.However,due to the problem of information asymmetry between enterprises and financial institutions,financial institutions usually set up many financing restrictions to avoid bad debt losses,which makes it difficult for small and medium-sized enterprises to carry out external financing and restricts the rapid development of enterprises.Therefore,there has been a financing method based on supply chain finance.In this process,the probability of credit risk in enterprises is also increasing.In this context,the construction of a comprehensive risk index system and risk assessment model has a certain role in promoting the stable operation and risk management of enterprises.The main contents of this thesis are as follows :(1)After introducing the relevant theoretical basis,this thesis discusses the source and selection criteria of credit risk evaluation index,and establishes the credit risk index system of small and medium-sized enterprises suitable for this thesis.The index system covers many influencing factors,and finally determines 5 first-level indicators,11 second-level indicators and 32 third-level indicators.Combining financial indicators with qualitative non-financial indicators will help to objectively evaluate the credit risk of financing enterprises.This thesis selects 307 manufacturing enterprises as samples.(2)in the process of establishing the credit risk assessment model of small and medium-sized enterprises,after establishing the Logistic regression model and the LSTM model,the whale optimization algorithm is set up with new parameters for optimization and added to the LSTM model to construct a combined model,and a risk assessment model suitable for this thesis is constructed.(3)In the experimental analysis part,the article quantifies the qualitative indicators and processes them together with the financial indicators.Factor analysis is used to screen out 10 principal component variables.Logistic regression model,neural network model and optimized neural network model are used to evaluate credit risk.The results show that the prediction accuracy of the optimized neural network model WOA-LSTM is higher than that of Logistic and LSTM models,and it is more suitable for the evaluation model of this thesis.Finally,this thesis puts forward some suggestions,such as improving the information sharing mechanism,constructing a complete and comprehensive risk index system and cultivating a diversified financial system,so as to reduce the probability of credit risk,which has a certain reference for financial institutions to improve their risk management level from the perspective of supply chain finance.