A Review of Enhanced Index Price Movement Prediction Using Ensemble Deep Learning Models

Abstract

This review paper investigates and evaluates advancements in index price movement prediction through ensemble deep learning models. Focusing on the fusion of Convolutional Neural Networks (CNN) and Deep Neural Networks (DNN), the study comprehensively analyses their collective impact on enhancing the precision and accuracy of index price forecasts in financial markets. The paper reviews and compares various ensemble strategies employed in deep learning, assessing their effectiveness in capturing complex patterns and dependencies within financial data. Special attention is given to the synergistic integration of CNN and DNN, highlighting how their combined capabilities contribute to improved predictive performance. Through an extensive literature review, the study examines the methodologies, architectures, and training strategies associated with ensemble deep learning models in the context of index price prediction. Additionally, the paper explores ensemble techniques’ implications for mitigating overfitting and enhancing model robustness in dynamic financial environments.