The Effects of Renewable Energy Consumption on Financial Performance: An Explainable Artificial Intelligence (XAI)-Based Research on the BIST Sustainability Index

Abstract

The main purpose of this study is to analyze the relation between renewable energy consumption and the financial performance of businesses included in the Borsa Istanbul (BIST) Sustainability Index using the Explainable Artificial Intelligence (XAI) method and to evaluate its predictability using machine learning methods. Within the scope of this research, annual business data from 2021 to 2023 are used. In this context, this study compared the forecast performances of different machine learning (ML) methods (XGBoost, Random Forest, and LightGBM) by analyzing them using SHAP (SHapley Additive Explanations), an XAI method. Some variables in this study consist of non-financial indicators, while the others consist of financial accounting and market-based indicators. The results show that renewable energy consumption has a very limited negative effect on ROA, whereas an increase in the total renewable energy ratio provides a positive contribution to ROA, albeit at a low level. Renewable energy consumption has a low positive effect on ROE. The XGBoost for ROA and Random Forest algorithm for ROE were more successful in estimation. This study goes beyond traditional statistical analyses and reveals the financial effects of accounting indicators in detail using XAI-based methods and contributes to enterprises’ decision-making processes regarding energy transformation. © 2025, Econjournals. All rights reserved.