Comparative Analysis of LSTM and Grid Search Optimized LSTM for Stock Prediction: Case Study of Africa Energy Corp. (AFE.V)

Mokona, Boho
Shipo, Ngezana

International Journal Artificial Intelligent and Informatics
Vol 2, No 1 (2024)

This research examines the effectiveness of Long Short-Term Memory (LSTM) neural networks for predicting Africa Energy Corp. (AFE.V) stock prices, comparing a standard LSTM implementation with a Grid Search optimized LSTM model. The research shows that hyperparameter optimization through Grid Search significantly improves prediction accuracy. The optimized LSTM model achieved superior performance across all evaluation metrics, with a test RMSE of 0.01, MAE of 0.01, MAPE of 3.41%, and R² of 0.9518, showing substantial improvement over the model without optimization. These findings emphasize the importance of hyperparameter tuning in deep learning models for financial time series forecasting and provide empirical evidence supporting the application of optimized LSTM networks for stock price prediction.