Machine learning models compared for Egyptian stock market forecasting

A new study published on arXiv analyzes the effectiveness of various machine learning models for forecasting the Egyptian Stock Exchange's EGX30 index. The research compares models like K-Nearest Neighbours, random forest, extreme gradient boosting, and recurrent neural networks (LSTM, GRU). Results indicate that the Gated Recurrent Unit (GRU) model performed best for one-week, one-month, and two-month predictions, while eXtreme Gradient Boosting (XGBoost) excelled in one-day forecasts. The study also highlighted the benefits of ensemble techniques for long-term predictions and noted the surprisingly strong performance of K-Nearest Neighbours in long-term forecasting. AI

IMPACT Provides insights into effective machine learning models for financial market prediction in developing economies.

RANK_REASON Academic paper analyzing machine learning models for financial forecasting. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.AI →

• EGX30
• Egyptian stock market
• Gated Recurrent Unit
• gated recurrent unit networks
• K-Nearest Neighbours
• Muhammed Ibrahim Walid
• random forest
• extreme gradient boosting

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →