Abstract: Large and medium-sized coal enterprises have the characteristics of high market relevance and large risk losses.Once credit risk occurs, the impact on the company and society will be huge.To accurately identify the credit risk of coal enterprises, this paper takes listed coal enterprises as the research object, and proposes a credit risk assessment model for large and medium-sized coal enterprises based on Filter-Wrapper two-stage feature selection.First of all, according to the characteristics of large and medium-sized coal enterprises, two new indicators are proposed in combination with the risk factors of coal enterprises in the selection of general indicators:anti-risk ability, gross profit margin of coal and processed products business sales; then the Filter-Wrapper two-stage feature selection algorithm is used to filter redundant features to build a credit risk prediction model.The experimental results show that the proposed model has higher prediction accuracy than before screening, and higher recognition rate of credit risk default samples.It verifies the effectiveness of the model and the proposed indicators, which is of great significance to the credit risk identification of large and medium-sized coal enterprises.