Framework For Price Prediction In The Nigerian Stock Market Using Data Mining Techniques
Hafsat Amina Jibril
Computer Science Department Modibbo Adama University, Yola, Adamawa State
Abdullahi Salisu Kadir
Computer Science Department Modibbo Adama University, Yola, Adamawa State
Abstract
The stock market has become an integral part of the modern financial system, playing a significant role in shaping the global economy. The stock market prices serve as a key indicator of market sentiment, economic performance and investor confidence, with a growing demand for accurate stock market predictions. This research explores a data mining framework for Nigerian stock market predictions using Decision Tree, Support Vector Regression, and Artificial Neural Network techniques. Data from Dangote Sugar Refinery is used to create new variables for models. The models are evaluated using MAPE and MSE, and show promising outcomes. The combined models have the highest accuracy at 85%, capturing both short-term and long-term trends effectively. The study recommends further development of advanced sentiment analysis and dynamic model adaptation for better implementation