| Affected by the trade war and the late stage of the epidemic,companies in China’s foreign trade industry have generally suffered a greater blow.Under the current difficulties in the foreign trade industry,it is particularly important to choose a group of foreign trade companies with low credit risk to provide loan assistance to drive the recovery and development of the entire industry.However,there is still a lack of an effective quantitative credit rating model for companies in the foreign trade industry.Therefore,it is necessary to establish a credit rating method for listed companies in the foreign trade industry in China.Firstly,the listed companies in the foreign trade industry taken as the research object,by analyzing the influencing factors of the credit rating of the listed companies in the foreign trade industry,a credit rating index system is constructed that includes the specific indicators of foreign trade specifically for the listed companies in the foreign trade industry;The system grades foreign trade listed companies according to whether they are ST stocks as the class label.In view of the small sample size of foreign trade listed companies and the imbalance of ST categories,a combination of ROSE and random forests are used.Rating classification method,carry out category balance processing and model classification rating for foreign trade listed companies;then,considering that the original ROSE method artificially synthesized too much data,it is easy to destroy the integrity of the original data,and innovatively proposes an improved ROSE method.On the basis of the classification,the data is balanced.Finally,comparing the original ROSE method and the improved ROSE method combined with the random forest rating model,it is found that the improved model has achieved great results in terms of accuracy and true negative rate.improved,and the model fit was better.Finally,based on the research on the credit rating of listed companies in China’s foreign trade industry,a reference for the future development direction is summarized and provided. |