Research On Risk Evaluation Of Small And Micro Enterprises' Loan Default

Small micro enterprise has always been an important part of our socialist market economy,playing an irreplaceable role in increasing employment,technological innovation and economic growth in China and paying tax.On the one hand,in the current Internet + and artificial intelligence in the era of large data,it opens up a small micro enterprise progressive development path.However,at the same time,the financing problem faced by small and micro enterprises has become one of the main reasons for their weak growth.On the other hand,since the reform of the banking industry,a large number of township banks have sprung up and intensifies the competition in the banking industry.What's more,their loan business for small and micro enterprises has become the main position for them to seize the market share.This paper builds a set of early warning models for credit risk of small and micro enterprises in banks.This paper will take the small and micro enterprise loan activities carried out by Guangzhou X bank as the research object,combined with the X line in Guangzhou.By using the individual sample data of Guangzhou X bank,selecting the loan amount and loan interest rate,loan term selection,classification,loans,loans to purposes,collateral loan type,and loan source as the basic data of the statistical analysis and build the cox model and logistic regression early warning model.After the inspection,we found that the warning effect of cox model is superior to logistic regression model in practical application.The following findings are obtained: in the collateral classification,small and micro business customers mortgaged by current assets are more likely to be overdue than small and micro business customers mortgaged by fixed assets;In terms of loan usage,the overdue probability of small and micro enterprise customers with working capital loan is higher than that of small and micro enterprise customers with fixed capital loan.In terms of loan investment,small and micro enterprises in real estate related industries are less likely to be overdue,while small and micro enterprises in real industry are more likely to be overdue.Among the types of loan occurrence,small and micro enterprise customers borrowing new loans are the most likely to be overdue,followed by small and micro enterprise customers borrowing new loans,and new small and micro enterprise customers borrowing old loans are the least likely to be overdue.In terms of loan sources,the delinquency rate of small and micro enterprise customers with online loans is significantly higher than that of small and micro enterprise customers with offline loans.Based on the above findings,targeted suggestions and measures are proposed for banks.When actively expanding the market of small and micro enterprises,banks should improve their risk control ability and pay attention to credit risk prevention.At the same time,it also provides reference for the control of credit risk of other commercial banks in China.