top of page

Unlocking the Power of AI in Stock Market Prediction: An Overview of Recent Research Trends

Source :


Machine learning has had a significant impact on the stock market and its prediction, as seen in the numerous studies that have been conducted in recent years. Weiwei Jianghas conducted a comprehensive review of 66 research papers in 2021 to understand the impact of machine learning on the stock market [1]. He analyzed various stock market indexes, different types of data, different types of input features, and different types of artificial intelligence techniques, including machine learning, deep learning, and reinforcement learning [2]. He concluded that deep learning has shown impressive performance in algorithmic trading applications and that reinforcement learning (RL) remains a major study area in quantitative trading [3].

In a related study, the authors provided an overview of deep reinforcement learning (DRL) as it relates to trading on financial markets and identified common problems and limitations associated with such methods [4]. Another study looked at the research that has been conducted in the field of stock market forecasting using methods based on computational intelligence, such as artificial neural networks, fuzzy logic, and genetic algorithms [5]. The study concluded that while DRL in stock trading has great application potential, the field is still in its early stages and there is a lack of experimental testing on real-time online trading platforms [6].

Yue Deng et al. developed a system that combines deep learning (DL) and reinforcement learning (RL) to make trade choices in the stock market [7]. Another study proposed a nested reinforcement learning technique that integrates RL in the decision-making process using deep reinforcement learning models [8]. Another study presented a hybrid time series decomposition stock index forecasting model that used LSTM models to extract abstract characteristics [9].

In another study, the authors used the continuous transfer entropy approach as a feature selection criterion to analyze the prediction ability of various factors on the direction of bitcoin's price [10]. Another study addressed the issue of stock market prediction by developing the RCSNet hybrid model, which combines linear and nonlinear models [11]. A study conducted between 2015 and 2019 examined 244 publicly listed businesses in the Dhaka stock market and found that the ensemble classifier outperformed all other models [12].

Adversarial reinforcement learning (ARL) has also been shown to be useful in making market-making agents that can handle changing market conditions. A study compared two traditional single-agent RL agents with ARL and found that ARL led to the emergence of risk-averse behavior and significant improvements in performance [13]. Another study showed how RL-based decision trees can be used to train autonomous agents that are aware of risks [14].

In conclusion, there have been numerous studies conducted in recent years that have explored the impact of machine learning and artificial intelligence on the stock market and its prediction. These studies have shown that deep learning and reinforcement learning have great potential in algorithmic trading, but the field is still in its early stages and there is a need for more experimental testing on real-time online trading platforms.

bottom of page