Research On Loan Default Risk Prediction Of Small And Micro Enterprises Based On Support Vector Machine

Small and micro enterprises are small in scale,so it is difficult to effectively counter various risks,and the uncertainty in the development process is strong.However,small and micro enterprises are an important part of China’s economy.The healthy development of small and micro enterprises contributes to the stable growth of China’s economy.Small and micro enterprises face high operational risks,while Rural Commercial Bank and other banks need to take certain measures to control risks when carrying out inclusive finance small and micro enterprise loan business,which is likely to promote the development of inclusive finance small and micro enterprise business in a longer term.This paper studies the prediction of loan default risk of small and micro enterprises in Rural Commercial Bank of China.First of all,analyze the theory and literature related to the default risk prediction of small and micro loans.Secondly,analyze the current situation and existing problems of S Rural Commercial Bank’s small and micro loan default risk.In recent years,the non-performing ratio of S Rural Commercial Bank’s small and micro loans is higher than the average non-performing ratio of inclusive finance’s small and micro enterprise loans,and shows a trend of gradual increase.The scoring indicators of the former Rural Commercial Bank are mainly designed from the financial related factors of enterprises,which may be difficult to effectively and comprehensively reflect the risks faced by small and micro enterprises.At present,the scoring method cannot effectively distinguish the default risk of small and micro enterprises.Third,based on the support vector machine,build a small and micro loan default risk prediction model.Fourth,collect the small and micro enterprise loan data of Rural Commercial Bank of China for empirical analysis,and verify the model effect through comparative analysis with different models.The support vector machine model constructed for the small and micro loan default risk prediction model has a good prediction effect on the test set;With ten fold cross calculation,the prediction effect of the support vector machine model fluctuates less,and the prediction is relatively stable,with little difference from the prediction using the test set,indicating that the model has a certain stability.At the same time,with the index set constructed in this paper,the existing index set is used to improve the precision by 0.0225,the recall by 0.0571,the accuracy by 0.0499,the true negative rate by0.0368,and the AUC by 0.0455.