Machine Learning Models Forecast Egyptian Stock Market Trends.

Research has focused on applying machine learning to predict stock prices within the Egyptian EGX30 market, a significant financial hub in the Middle East. This area has seen less attention compared to global stock markets, despite the growing need to understand trends in developing economies. The study evaluated several machine learning models, including K-Nearest Neighbors (KNN), Random Forest, XGBoost, Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) networks. These models were assessed for their ability to forecast both short-term and long-term EGX30 trends using historical data and standard financial metrics. Results indicated that GRU models were most effective for predictions spanning one week, one month, and two months. For very short-term, one-day predictions, XGBoost demonstrated superior performance. Interestingly, ensemble techniques showed significant improvements for long-term forecasts, and KNN also performed surprisingly well in long-term scenarios, suggesting its continued relevance in fintech applications.