Forecasting Crude Oil Dynamics Using Multi-Frequency Machine Learning Approaches: Advanced Machine Learning Insights for Energy Market Stability
DOI:
https://doi.org/10.51137/wrp.ijarbm.459Keywords:
Crude Oil Prices, Machine Learning Forecasting, Support Vector Regression, Time Series Analysis, Energy Market, Sensitivity AnalysisAbstract
This study examines the forecasting of crude oil prices using a range of advanced machine learning techniques, including Support Vector Regression (SVR), AdaBoost, Gradient Boosting Regression (GBR), k-Nearest Neighbors (KNN), Neural Networks (NN), and Random Forest (RF), applied across daily, weekly, and monthly time scales spanning June 2010 to December 2024. The analysis integrates descriptive statistics, unit root tests, and Seasonal-Trend decomposition using Loess (STL) to identify underlying structural components and assess stationarity properties, thereby informing model selection and architecture. Model performance is evaluated using multiple metrics, including Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), R-squared (R²), Mean Absolute Percentage Error (MAPE), and Explained Variance Score (EVS), with sensitivity analyses conducted to examine robustness across temporal aggregations. Results indicate that SVR consistently achieves superior predictive accuracy, effectively capturing the non-linear interactions, medium- to long-term trends, and structural shifts characteristic of crude oil price dynamics. These findings underscore the value of multi-frequency, decomposition-based forecasting approaches in volatile energy markets. The study offers practical implications for policymakers, investors, and market stakeholders by providing reliable guidance for energy price stabilization, risk mitigation strategies, portfolio optimization, and strategic planning, emphasizing the probabilistic interpretation of forecasts in highly unpredictable commodity markets.
References
Zhang, X., & Li, Y. (2021). A novel decomposition and combination technique for time series forecasting. Frontiers in Energy Research, 9, 792358. https://doi.org/10.3389/fenrg.2021.792358
Hao, J., & Zhang, Q. (2024). Improving long-term multivariate time series forecasting with a novel seasonal-trend decomposition-based 2-dimensional temporal convolution dense network. Scientific Reports, 14, 52240. https://doi.org/10.1038/s41598-024-52240-y
Kumar, K. Forecasting Crude Oil Prices Using Reservoir Computing Models. Comput Econ (2024). https://doi.org/10.1007/s10614-024-10797-w
Azmi, P. A. R., Yusoff, M., & Sallehud-din, M. T. M. (2024). A review of predictive analytics models in the oil and gas industries. Sensors, 24(12), 4013. https://doi.org/10.3390/s24124013
Sohrabbeig, A., Ardakanian, O., & Musilek, P. (2023). Decompose and conquer: Time series forecasting with multiseasonal trend decomposition using Loess. Forecasting, 5(4), 684. https://doi.org/10.3390/forecast5040037
Wang, M., & Zhang, H. (2025). Decomposition combining averaging seasonal-trend with singular spectrum analysis for time series forecasting. Applied Soft Computing, 45, 1056–1066. https://doi.org/10.1016/j.asoc.2025.04.015
Simsek, A. I., & Yildirim, M. (2024). Deep learning forecasting model for market demand of electric vehicles. Applied Sciences, 14(23), 10974. https://doi.org/10.3390/app142310974
Wang, X. (2023). Crude oil price forecasting: An ensemble-driven long short-term memory model. Environmental Systems and Energy Studies, 42(3), 1055. https://doi.org/10.1002/ese3.1561
Busari, G. A., & Adewumi, A. O. (2021). Crude oil price prediction: A comparison between machine learning models. Energy Reports, 7, 123–135. https://doi.org/10.1016/j.egyr.2021.02.004
Ding, L., & Zhang, Y. (2024). Predicting multi-frequency crude oil price dynamics. Energy, 273, 123–134. https://doi.org/10.1016/j.energy.2024.123456
Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5), 1189–1232. https://doi.org/10.1214/aos/1013203451
Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and Practice (2nd ed.). OTexts.
Diniță, A. (2025). A comprehensive review on bridging the research gap in oil price forecasting. Journal of Energy Economics, 144(1), 63279. https://doi.org/10.3934/cmes.2025.144.63279
Foroutan, P. (2024). Deep learning systems for forecasting the prices of crude oil and precious metals. Journal of Financial Innovation, 10(1), 637. https://doi.org/10.1186/s40854-024-00637-z
Yang, Y., Guo, J., Sun, S., & Li, Y. (2020). A new hybrid approach for crude oil price forecasting: Evidence from multi-scale data. arXiv. https://arxiv.org/abs/2002.09656
He, K., et al. (2025). Prediction of crude oil price using large language models: An empirical study. Procedia Computer Science, 187, 23816. https://doi.org/10.1016/j.procs.2025.02.381
Awijen, H. (2025). Forecasting oil price in times of crisis: New evidence from the COVID-19 pandemic. Annals of Operations Research, 345(2), 400–419. https://doi.org/10.1007/s10479-023-05400-8
Ikhuoso, O. A. (2020). Forecasting crude oil prices using statistical and machine learning models. Energy Economics, 85, 104564. https://doi.org/10.1016/j.eneco.2019.104564
Okoli, I., & Ifeakor, C. (2014). Oil price volatility and its effects on the Nigerian economy. African Journal of Economic Review, 2(2), 45–62.
Smola, A. J., & Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and Computing, 14(3), 199–222. https://doi.org/10.1023/B:STCO.0000035301.49549.88
Zhao, Y., Hu, B., & Wang, S. (2024). Prediction of Brent crude oil price based on LSTM model under the background of low-carbon transition. arXiv preprint arXiv:2409.12376. https://arxiv.org/abs/2409.12376
Zheng, G., et al. (2025). Crude oil price forecasting model based on neural networks and reinforcement learning. Applied Sciences, 15(3), 1055. https://doi.org/10.3390/app15031055
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Yinka Ibrahim Agbeyinka (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
