<![CDATA[This study aims to examine the effect of machine learning disclosure and intellectual capital on stock returns through return on equity (ROE) in energy sector companies listed on the Indonesia Stock Exchange during the 2020–2024 period. Advances in data-driven technologies and increasing business competition have encouraged companies to adopt machine learning and effectively manage intellectual capital to improve corporate performance, reduce information asymmetry, and support decision-making processes aimed at minimizing business and investment risks. This study employed a quantitative approach using panel data from 51 energy sector companies selected through purposive sampling, resulting in a total of 255 observations. Machine learning disclosure was measured using a disclosure index developed through text mining of the companies' annual reports, while intellectual capital was measured using the Value-Added Intellectual Coefficient (VAIC™) method. The data were analyzed using panel data regression through the selection of the most appropriate estimation model, followed by estimation using the Feasible Generalized Least Squares (FGLS) method. Mediation analysis was conducted using Stata software. The findings indicate that machine learning disclosure has a positive but statistically insignificant effect on stock returns, while it has a positive and statistically insignificant effect on return on equity. Intellectual capital has a positive and statistically significant effect on both return on equity and stock returns. Furthermore, return on equity has a positive and statistically significant effect on stock returns. The mediation analysis reveals that return on equity does not mediate the relationship between machine learning disclosure and stock returns. However, return on equity partially mediates the relationship between intellectual capital and stock returns at the 10% significance level. These findings suggest that intellectual capital plays a crucial role in enhancing corporate profitability and stock returns, whereas machine learning disclosure remains limited in its ability to influence stock returns among energy sector companies during the study period.]]>
Copyrights © 2026
