Systematic Literature Review For Stock Price Prediction with Machine Learning

This systematic literature review assesses stock price prediction methodologies, concentrating on research employing machine learning, time series analysis, and statistical models. Following the implementation of rigorous inclusion and exclusion criteria, we examine research according to their aims, techniques, datasets, and outcomes. The comprehensive literature assessment of these 10 studies uncovers significant insights and consequences for the three study issues. The paper highlights the efficacy of AI models, especially in stock market prediction, utilizing Long Short-Term Memory (LSTM) networks due to their ability to process sequential data and recognize long-term relationships. Sentiment analysis derived from intermarket data, news, and social media boosts predicted accuracy, particularly when integrated with conventional technical analysis. Market variables, including historical returns, momentum, implied volatility (VIX), and cash flow financial data, enhance predictive accuracy by offering further context. Standard assessment indicators, such as accuracy, return on investment (ROI), profitability, and Mean Absolute Error (MAE), indicate that AI models surpass conventional techniques. Nonetheless, obstacles persist, including dependence on erratic sentiment data, dangers of overfitting, and possible biases arising from unstructured sentiment data and the limited volume of sentiment data. Future research may increase sentiment analysis by including additional sentiment data and implementing quality control, integrating diverse AI methodologies, and encompassing a wider range of financial and economic data to strengthen model robustness. This article seeks to emphasize the growing capability of deep learning to enhance stock market prediction methodologies.