企业 ESG 表现的多维驱动机制与提升策略OACHSSCD

Under the guidance of the new development concept,incorporating environmental,social responsibility,and corporate governance(ESG)into corporate operations and decision-making is an effective way to achieve sustainable development and Chinese-style modernization.Based on machine learning methods(ensemble learning and reinforcement learning),this paper uses 2009～2024 Chinese A-share listed companies as research samples and confirms the superiority of machine learning in model interpretability and prediction accuracy,analyzes the external pressures and internal governance motivations that affect corporate ESG performance,and comprehensively examines the multidimensional factors that predict corporate ESG,and the optimal strate-gies for improving corporate ESG are analyzed from the perspective of external pressure and internal governance.The results indicate that:first,both external factors(institutional environment,media attention,and industry competition)and internal governance factors(e.g.,ownership characteristics)impose significantly positive effects on corporate ESG performance.Second,external at-tributes including industry competition pressure and positive media coverage,together with internal indicators such as managerial ownership and the ownership ratio of the top five shareholders,serve as core predictors of ESG performance,with differentiated im-portance and nonlinear patterns.Third,heterogeneity analysis reveals that internal features(ownership,executive characteristics)and external factors(industry,institution,regional rule of law,government intervention)have heterogeneous impacts on different ESG dimensions and firm types.Finally,reinforcement learning-based strategy simulation suggests that maintaining moderate in-dustry competition,enhancing external monitoring via media,retaining relatively high managerial ownership and appropriate owner-ship concentration can maximize the improvement of corporate ESG performance.This study provides empirical implications for firms to design ESG optimization schemes and for authorities to improve relevant regulatory policies.