Published Paper Details:

Stock market prediction is one of the toughest tasks in finance and attracts much attention from academia and industry due to huge economic potential. Traditional econometric models dominated the field until recent developments in machine learning introduced new methodologies that capture complex market dynamics. Of these, the RF algorithm has excelled in large data volumes, the ability to model nonlinear relations, and ranking feature variable importance. The paper presents a review on the application of Random Forests in stock market prediction, covering the theoretical foundations of Random Forest, advantages over traditional methods, reviews of empirical studies proving their effectiveness, and challenges and future research directions. It synthesizes how recent literature can help researchers and practitioners use random forests to guide and improve the accuracy of predictions and robustness associated with stock market analysis.