At present, the correlation between environmental performance and financial performance has become a key issue in corporate sustainable development strategies. However, traditional statistical analysis methods have certain limitations in dealing with such multivariate relationships, especially when the data dimension is high and the complex relationships between variables are strong. To address this problem, this study proposes a method based on the random forest algorithm to analyze the intrinsic connection between the environment and finance. First, by preprocessing the enterprise data. Then, the RF algorithm is used to model and analyze the data, which can effectively capture the complex nonlinear relationship between environmental performance and financial performance. Finally, through feature importance analysis, the environmental factors that have the greatest impact on financial performance are further identified. The experimental results show that the mean squared error (MSE) of the random forest model is only 0.042, which is significantly lower than ridge regression and linear regression. In addition, feature importance analysis shows that factors such as carbon emissions (impact weight 0.285) are the most critical environmental factors affecting financial performance. The above data conclusions show that when optimizing financial performance, companies should give priority to carbon emission management and energy efficiency improvement, while strengthening waste management and water resource utilization efficiency.