Do CEO and chairman characteristics affect green innovation? evidence from a comparative analysis of machine learning models

Using the extreme gradient boosting (XGBoost) algorithm, which is at the forefront of machine learning algorithms, this study comprehensively examines the impact of CEO and chairman characteristics on corporate green innovation. It has been used a sample of listed companies in China from 2010 to 2022 to compare it with the gradient-boosted decision tree (GBDT) model and multiple linear regression (MLR) model. It has been found that (1) the characteristics of the CEO and chairman of the board of directors of companies have a weaker predictive ability for corporate green innovation; (2) among the many personal characteristics of CEO and chairman, duality and age have a stronger predictive ability for corporate green innovation; (3) in addition to CEO duality, the relationship between age, environmental awareness, and green innovation have been characterised by nonlinearity, which is more in line with previous theories; (4) compared to the MLR and GBDT models, the XGBoost model has a higher prediction accuracy, with good performance in terms of goodness of fit, mean error, and mean square error. It is the first time, this study has examined the drivers of green innovation from a broader perspective using machine learning methods and it also provides useful insights for CEO and chairman appointments, incentive design, and sustainable corporate development.

Text 115658.pdf - Published Version Restricted to Repository staff only Download (2MB) Official URL or Download Paper: https://link.springer.com/article/10.1007/s10668-0...

Actions (login required)

View Item