Comparative analysis of machine learning models for stock price analysis across different dataset sizes

Abstract

The nature of the stock market has always been ambiguous as it constantly fluctuates for various factors. The regular fluctuations have always made it difficult for investors to invest. This paper compares six different machine learning models for stock price analysis: Explainable AI, Q-learning method, LSTM, Bi-LSTM, Restricted Boltzmann Machine, and Deep Belief Network. Each model was evaluated using three different datasets consisting of 7000, 10000, and 14000 data points, respectively. The results of the experiments show that depending on the size of the dataset, the performance varies and the specific model used. In general, the deep learning models (LSTM, Bi-LSTM, Restricted Boltzmann Machine, and Deep Belief Network) outperformed the Explainable AI and Q-learning models in terms of predictive accuracy. However, the Explainable AI and Q-learning models had the advantage of being more interpretable and easier to understand, which may be desirable in certain applications. Overall, this study provides insights into the strengths and weaknesses of various machine learning models for stock price analysis and highlights the importance of choosing the right model for the specific task at hand. Future work concentrates on optimizing the performance of the models further or exploring the use of hybrid models that combine the strengths of multiple approaches.