Predicting BELEX15 Stock Index Movements Using Artificial Neural Networks

Prediction of stock price index direction is a challenging task due to the complex and dynamic nature of financial markets. Accurate forecasts can yield substantial benefits for investors. This study develops an artificial neural network (ANN) based model to predict next-day movements of the BELEX15 Index, an emerging market index of the Belgrade Stock Exchange. Eleven technical indicators were selected as input features. A series of parameter-setting experiments were performed to optimize the ANN architecture, including hidden neurons, training functions, and learning parameters. The model was trained and validated using daily data from 2006 to 2024. The proposed model achieved 71.39% accuracy, outperforming comparable ANN models applied to other stock markets and significantly exceeding attempts to apply existing models to BELEX15. These findings demonstrate the potential of ANN-based models for market-specific forecasting and their utility in designing effective trading strategies in emerging financial markets.