<![CDATA[The realm of stock market forecasting presents a formidable challenge, given the intricate, noisy, chaotic, and ever-evolving nature of its time series data. However, the advent of computational advancements offers a ray of hope, as intelligent models hold the potential to assist investors and analysts in mitigating the inherent risks associated with financial markets. In recent years, Deep Learning models have garnered significant attention, with numerous studies delving into their application for predicting stock prices using historical data and technical indicator Yet, the ultimate goal in this pursuit is not merely prediction but validation, a crucial step in the context of the financial market. This systematic review sets its sights on Deep Learning models employed in stock market forecasting through the lens of technical analysis. It dissects the landscape based on four pivotal dimensions: predictor techniques, trading strategies, profitability metrics, and risk management. Unveiling the findings, it becomes apparent that the LSTM (Long Short-Term Memory) technique reigns supreme, representing a substantial 73.5% of the studies in this domain. However, the review uncovers notable limitations in the existing literature, with a mere 35.3% of studies addressing profitability metrics and a mere two articles delving into the intricacies of risk management.]]>
Copyrights © 2023
