A Hybrid Approach for Stock Market Price Forecasting Using Long Short-Term Memory and Seahorse Optimization Algorithm

Abstract Stock market price forecasting is a challenging task due to the complexity and volatility of market dynamics. This paper proposes a novel approach that combines the strengths of Long Short-Term Memory (LSTM) networks and the Seahorse Optimization (SHO) algorithm for stock market price forecasting. The LSTM-SHO model is compared with other LSTM models optimized using Genetic Algorithm (GA) and different dimensionalities (1D, 2D, 3D), as well as an Artificial Neural Network (ANN) model. The results show that the LSTM-SHO model outperforms the other models in terms of mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (R2). The hyperparameter tuning process using SHO significantly improves the forecasting accuracy of the LSTM model. The proposed approach demonstrates its potential in developing more accurate and robust stock market forecasting models, which can aid investors and analysts in making informed decisions. The findings of this study have important implications for investors, analysts, and policymakers, and contribute to the existing body of literature on stock market forecasting.