International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 186 - Number 74 |
Year of Publication: 2025 |
Authors: Anika Tahsin Biva, A.B.M. Shahadat Hossain, Md. Shafiul Alom Khan, Iqbal Habib |
![]() |
Anika Tahsin Biva, A.B.M. Shahadat Hossain, Md. Shafiul Alom Khan, Iqbal Habib . Benchmarking Hybrid ANN-LSTM and Physics-Informed Neural Networks for Forecasting Stock Market Prices. International Journal of Computer Applications. 186, 74 ( Mar 2025), 1-8. DOI=10.5120/ijca2025924624
This study presents a novel approach for forecasting stock market prices by combining Artificial Neural Networks (ANN) and Long Short-Term Memory (LSTM) models into a hybrid ANNLSTM framework. This study focus on forecasting the closing prices of the S&P 500 and Toronto Stock Exchange (TSX) indices, evaluating the performance of the proposed hybrid model against traditional ANN, LSTM, and Physics-Informed Neural Network (PINN) models. The hybrid ANN-LSTM model demonstrates superior forecasting accuracy, outperforming the individual models and the PINN in terms of multiple evaluation metrics. The training dataset spans from January 1, 2005, to December 31, 2020, while the testing period covers January 1, 2021, to January 31, 2024. The results highlight the potential of hybrid deep learning models, specifically the ANN-LSTM combination, in enhancing stock market prediction accuracy, representing a significant advancement over conventional methods.