A Financial Early Warning Analysis Model for Enterprises Based on the Algorithm of Composite Minority Oversampling Technology
Financial crisis can lead to difficulties in capital turnover for enterprises,and the inability to connect production and sales processes,thereby reducing the profitability of the enterprise.Therefore,in order to achieve accurate prediction of financial crises and remind enterprises to take timely response measures,a financial crisis evaluation index system has been studied and established,and a crisis prediction model based on synthetic minority class oversampling technology algorithm and random forest has been proposed.The experimental results showed that compared to random forests and support vector machines,the crisis prediction model proposed in the study predicted 17 ST enterprises in the test set,and there were no false positives.The average accuracy,average accuracy,and area under the receiver operating characteristic curve of the model are all higher than other models,with values of 97.2%,96.7%,and 0.91,respectively.From this,it can be seen that the crisis prediction model proposed in the study can accurately predict the financial crisis of enterprises and provide accurate financial warnings for them.
万方数据
维普
