Scholarlink

<p>The research examines how board gender diversity, as well as innovation and environmental, social, and governance practices, affect environmental, social, and governance (ESG) performance through machine learning techniques. The study adopts Linear Regression and Random Forest algorithm to explore complex relationships between ESG drivers between 2005 and 2024 in Germany. Social and governance scores significantly impact ESG performance; however, machine learning analysis reveals the intricate relationship between environmental innovation and gender diversity. The Random Forest model surpasses linear and ridge regression regarding accuracy and generalizability. The study provides evidence for stakeholder and upper echelon theories by demonstrating how diverse boards and sustainable innovation contribute to positive ESG outcomes. It extends ESG research by applying machine learning, which offers strong alternatives to conventional econometric analysis while delivering operational guidance for firms seeking sustainability improvements through governance and innovation measures. The key contribution of this study lies in its methodological advancement, integrating interpretable machine learning techniques to uncover nonlinear and interaction-driven ESG dynamics, while offering empirical insights into how innovation capabilities and board diversity jointly shape corporate sustainability performance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine learning insights into ESG performance: the influence of innovation and gender diversity

  • 期刊论文

  • 发表日期：

    2025年9月19日

查看全文
  • Kemal Cek,
  • Demet Beton Kalmaz

摘要

The research examines how board gender diversity, as well as innovation and environmental, social, and governance practices, affect environmental, social, and governance (ESG) performance through machine learning techniques. The study adopts Linear Regression and Random Forest algorithm to explore complex relationships between ESG drivers between 2005 and 2024 in Germany. Social and governance scores significantly impact ESG performance; however, machine learning analysis reveals the intricate relationship between environmental innovation and gender diversity. The Random Forest model surpasses linear and ridge regression regarding accuracy and generalizability. The study provides evidence for stakeholder and upper echelon theories by demonstrating how diverse boards and sustainable innovation contribute to positive ESG outcomes. It extends ESG research by applying machine learning, which offers strong alternatives to conventional econometric analysis while delivering operational guidance for firms seeking sustainability improvements through governance and innovation measures. The key contribution of this study lies in its methodological advancement, integrating interpretable machine learning techniques to uncover nonlinear and interaction-driven ESG dynamics, while offering empirical insights into how innovation capabilities and board diversity jointly shape corporate sustainability performance.