Profitability is a crucial measure of financial stability and operational success for firms. In Indonesia, the capital market has grown significantly, with the Indonesia Stock Exchange (IDX) reaching a market capitalization of IDR 11.67 quadrillion by 2023. However, there remains a gap in studies that comprehensively analyze the determinants of profitability across all non-financial sectors in Indonesia. This research aims to identify and analyze the determinants of profitability in Indonesian non-financial companies using both traditional panel data regression and machine learning techniques. Using quarterly data from 816 non-financial companies listed on the IDX from 2012 to 2023, this study employs panel regression with a fixed effects model and Driscoll-Kraay standard errors. Return on assets (ROA) and earnings per share (EPS) are employed as profitability measures, while firm size (LSIZE), company efficiency (CE), liquidity (LIQ), market power (MP), sales growth (SG), and sustainable growth rate (LSGR) are investigated as explanatory variables. Results from the panel regression analysis reveal that, except for LIQ, all variables have a positive and significant impact on profitability. The analysis is further refined using machine learning techniques, specifically Random Forest, XGBoost, and a deep learning neural network, which conclude that the most important variable influencing ROA is company efficiency, while the most important variable influencing EPS is firm size
Copyrights © 2025