Research On Credit Risk Evaluation Of Small And Medium Enterprises Under Supply Chain Finance

In recent years,the financial credit market is booming in China.The term credit risk has gradually become well-known and concerned by the public.Credit rating agencies emerge in an endless stream and their position in the market is also rising steadily.Foreign credit rating system has formed a scale,meanwhile,domestic research on this aspect started late and is not systematic.However,the consequent supply chain financial service model has accelerated the pace of research in this field in China.Domestic scholars and experts have been committed to exploring credit risk level rating methods in line with China's national conditions and market economy system.Summarizing the above issues and taking them as the starting point and objective of the research,this article makes use of SPSS20.0 statistical software and Stata14.0 statistical software to statistically analyze the relevant data indicators of SMEs' credit risk in supply chain finance.The factors influencing the credit risk level of SMEs are multiple and complex,among which eight third-level indicators have a negative correlation with the credit risk rating level of SMEs.Combined with the results of statistical experiments,a rational solution to the problem of SMEs' credit risk in supply chain finance is proposed from the perspective of SMEs,commercial banks and other financial institutions.Summarizing the formation mechanism of credit risk on the supply chain financial chain and SMEs,this article highlights the Mixed-Logistic statistical regression model and improves the selection of credit risk evaluation indicators for SMEs.Adopting the method combined discrete numerical quantitative indicators with non-numerical qualitative indicators to measure the extent of credit default of SMEs,By constructing the Mixed-Logistic statistical model,60 sample data were finally screened and the model's discrimination rate of sample data is as high as 98%.In addition,150 experimental data were selected into the statistical model for testing.The statistical results show that the constructed statistical model can effectively distinguish between defaulting enterprises and normal enterprises.Figure 8;Table 13;Reference 54...