Research On Credit Evaluation Of Small And Medium-sized Enterprises In China

Small and medium-sized enterprises(SMEs)are an indispensable part of the development of China's national economy and a fundamental force to promote the development of national economy and social stability.At the end of 2018,there were369,000 small and medium enterprises,according to National Bureau of Statistics of the People's Republic of China.Among them,50,000 are medium-sized enterprises and 319,000 are small-sized enterprises.SMEs have short life span and high credit risk,so large number of SMEs make it urgent to use an effective credit evaluation model.At present,foreign scholars mostly take domestic historical data as samples to develop credit rating models for SMEs,and the applicability of design ideas for Chinese enterprises and the fit degree of model parameters for China's economic environment are worth discussing.However,early domestic scholars lacked data sources and used qualitative methods such as analytic hierarchy process(ahp)in their research methods,lacking data support.Most recent studies have taken the credit data of commercial Banks as samples,and the enterprise samples have been audited by Banks for risk control,and their represent ativeness is relatively weak.At present,NEEQ enterprises is more suitable as the representative of SMEs because of the increase of sample quantity,the open and audited financial information and so on.Considering that different industries have different operation and essence,which will lead to different research results,this paper selects the most representative information technology industry of NEEQ as the sample,a total of 2470 enterprises,the 59 ST and ST* Enterprises were classified as those with credit risk.First,the data were processed with missing value,abnormal value and dimensionless,then Smote algorithm was used to solve the sample imbalance problem.Based on the analysis of debt-paying ability,capital structure,operating ability,profitability,development ability and per-share index,this paper selects 27 characteristic variables and selects them by factor analysis.Then,the credit of small and medium-sized enterprises is evaluated by Logistic regression in statistical analysis method and stochastic forest in data mining technique,and AUC and stability coefficient as evaluation criteria,the random forest was optimized by optimizing the parameters,and the accuracy of the model was improved.Finally,this paper puts forward some suggestions on the credit rating of SMEs by commercial banks and the stable development of SMEs.The research of this paper is of great significance to identify the defaulting enterprises,improve the financing efficiency of SMEs,and determine whether the banks grant loans to SMEs.Through the construction of NEEQ information technology industry default model,to help investors better identify the default risk of SMEs.