A Study On The Credit Evaluation Methodology Of SME Financing In Shaanxi Province

With the pace of economic development,SMEs play a non-negligible role in promoting China’s GDP.However,the sudden arrival of the epidemic has hit many SMEs,and problems such as financing difficulties have drawn the attention of the government,commercial banks,SMEs and other practical circles as well as academics.How to solve this problem and improve the general environment of financing is an important task that enterprises themselves,banks and governments are urgently facing at present.First of all,this paper studies the development of credit evaluation system,addresses the lack of consideration for factors such as management ability of enterprise managers,enterprise industry prospect and quality of managers,and the unsoundness of evaluation index system,and based on the characteristics of incomplete financial data and more qualitative indexes of SMEs,makes up for the lack of subjectivity of expert judgment method by using the advantages of financial ratio analysis method,and integrates the advantages of expert judgment method and financial ratio analysis method.The credit risk evaluation system of SMEs in Shaanxi Province is constructed by using the advantages of the financial ratio analysis method and the comprehensive financial ratio analysis method.The model construction uses bp neural network model,SVM support vector machine model,and logistics regression model for comprehensive comparison and evaluation.Secondly,this paper combines the characteristics of SMEs in Shaanxi Province itself in the index selection,considering ten aspects of enterprise own strength,enterprise prospect,supply and sales relationship,managerial quality,management level,credit situation,operation capacity,growth capacity,solvency,and profitability,using 115 SMEs data in Shaanxi Province in 2021 as the research sample,13 financial indicators were selected,and 20 non-financial indicators as primary screening indicators,of which 6 non-financial indicators were evaluated by the bank’s internal experts.Significance tests were conducted,and eight principal components were selected as independent variables in combination with principal component analysis,and whether default was taken as the dependent variable Y.SVM,BP neural network,and logistic credit risk evaluation models were established.selected as independent variables in combination with principal component analysis,and whether default was taken as the dependent variable Y.SVM,BP neural network,and logistic credit risk evaluation models were established.Finally,through comparative analysis,it is concluded that the logistic model is superior to the SVM model and BP neural network model from the perspective of overall evaluation accuracy.The findings of this paper provide some reference significance for the credit risk evaluation of SMEs in Shaanxi Province,and hopefully provide a small help for commercial banks to better carry out related business.