Research On Early Warning Of Debt Repayment Risk Of Small And Medium-sized Enterprises Based On Neural Network

After the financing difficulty of SMEs has become a hot issue in China,many scholars have explored ways and means to solve the financing difficulty of SMEs from the perspective of the solvency of enterprises.The early warning technology of debt repayment risk is a reference for objective evaluation of SMEs'debt repayment ability.The early warning of SMEs'debt repayment risk status is the premise of judging the debt repayment risk level and the starting point of debt repayment risk management and control.Through a large number of literature reading,it is found that there are few studies on enterprise risk early warning from the perspective of debt repayment risk at home and abroad,and there is a lack of quantitative research on SMEs debt repayment risk early warning.Therefore,this paper firstly analyses the current situation of SMEs'debt repayment risk listed in China,and then constructs the early warning index system of SMEs' debt repayment risk combined with literature research.Then,the BP neural network model is used to warn the debt-paying risk of small and medium-sized enterprises in China.Subsequently,the decisive indicators obtained from the analysis of early warning results are tested.The following three conclusions are drawn:firstly,the SMEs'repayment risk model constructed in this paper can maximize the accuracy of early warning of SMEs' repayment risk;secondly,by comparing with the large enterprises'repayment risk early warning,the SMEs' repayment risk model constructed in this paper can provide a more scientific early warning method for SMEs'repayment risk.Thirdly,the analysis of the results of this model shows that the return on net assets is the most important index affecting the prediction accuracy of the early warning model of SMEs'debt repayment risk in China.Finally,it puts forward corresponding suggestions for SMEs to reduce the risk of debt repayment.