Machine Learning Algorithms to Study the Impact of Sustainability on Financial Success: Evidence from US Stock Market

In recent years, as in almost every field, the concept of sustainability has become a concept that has been given considerable importance and researched in both the academic and business worlds in the field of finance. This study aims to find the best model for estimating financial success within the framework of sustainability by adding environmental risk score, social risk score, and governance risk score in addition to classical indicators in stock returns. For this purpose, data obtained from 300 US companies across 11 distinct industries is utilized. The features evaluated are the environmental risk score, social risk score, governance risk score, as well as profitability, liquidity, leverage, RoA, and beta, which are used in classical studies. To obtain the best modeling predictions, various machine learning algorithms were used instead of classical statistical methods. Based on the F1 performance metrics of the seven machine learning algorithms tested, the model with the highest performance is Random Forest, an ensemble learning model. Based on the Random Forest model, environmental and social risk scores are particularly important features for financial success. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.