MACHINE LEARNING AND STOCK RETURN PREDICTION: EVIDENCE FROM THE NIGERIAN EMERGING MARKETS
DOI:
https://doi.org/10.57233/ijamer.v2i2.41Keywords:
Machine Learning, Stock Return Prediction, Emerging Markets, Asset Pricing, Financial EconometricsAbstract
This study investigates the effectiveness of machine learning techniques in predicting stock returns in the Nigerian stock market. Using daily stock prices from 2015 to 2024, the study examines the predictive power of technical indicators (RSI, MACD, Moving Averages, Momentum), market variables (trading volume, market returns, volatility), and macroeconomic factors (interest rate, inflation, exchange rate). Machine learning models, including Artificial Neural Networks (ANN), Support Vector Machines (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost), were implemented with hyperparameter tuning and cross-validation to ensure robustness. Results indicate that technical indicators are the most influential predictors, with RSI and MACD showing the highest importance. Ensemble models, particularly XGBoost, achieved superior predictive accuracy, outperforming single models in out-of-sample tests. Market variables contributed moderately, while macroeconomic factors played a secondary role in short-term prediction. The study concludes that combining technical indicators with advanced machine learning models provides a reliable framework for forecasting stock returns in volatile emerging markets like Nigeria. The findings offer practical insights for investors, portfolio managers, and policymakers seeking to improve decision-making using data-driven approaches.
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Author(s)

This work is licensed under a Creative Commons Attribution 4.0 International License.
Google Scholar
CORE
BASE (Bielefeld Academic Search Engine)