An advanced and comprehensive hybrid approach for analyzing crude oil futures prices

Abstract

Fossil fuels, such as crude oil, play a critical role in the global energy supply chain. The prices of crude oil significantly impact various businesses and economies worldwide. Predicting crude oil prices is particularly challenging in oil-producing countries due to the nonlinear and dynamic nature of price fluctuations, which are characterized by significant uncertainty. This study aims to examine factors that influence WTI crude oil prices over an extended period, from June 2014 to October 2023. Understanding the fundamental issues within the crude oil market is crucial for this analysis. Given the inherent volatility of crude oil prices, which arises from a range of factors, advanced analytical techniques are required to improve forecasting accuracy. To address this challenge, the study employs the Radial Basis Function (RBF) model, an effective ensemble learning technique renowned for its capability to identify complex patterns in datasets. The RBF model is enhanced with three optimization strategies to boost its predictive accuracy and optimize its parameters: Zebra Optimization, Barnacles Mating Optimization, and Fruit-Fly Optimization. The research findings suggest that integrating optimization techniques, particularly the Zebra Optimization Algorithm, with the RBF model can significantly enhance the accuracy of crude oil price forecasts. The proposed model achieved a coefficient of determination (R²) of 0.9889, indicating a high level of explanatory power. This result demonstrates the model's ability to accurately capture fluctuations in West Texas Intermediate (WTI) crude oil prices by incorporating trading volume as well as open, high, low, and close prices.