This study explores the use case of Long Short-Term Memory (LSTM) deep learning network architecture applied to stock market price forecasting. Long Short-Term Memory models, known for their ability to capture long-term dependencies in sequential data, are particularly suited for handling the non-linear and dynamic relations present in financial time series. For this case study, a model has been trained using Apple’s historical data of daily transactional prices and volumes, two technical indicators used in trading strategies, and macroeconomic data for the United States of America. After testing several architectures using the Bayesian-based search method, the results indicate that LSTM models can effectively model complex market behavior using the right architecture for each use case. This work contributes to the evolving field of financial analytics using modern data science methods by highlighting the potential of deep learning approaches in the volatile context of stock markets.