Answers are grounded in this paper and cite their sources.
Crude oil is a cornerstone of the global economy, with profound economic, geopolitical, and environmental implications. However, forecasting world crude oil consumption is inherently complex due to the dynamic interplay of multiple influencing factors. This study presents a novel and robust modeling approach aimed at accurately predicting global crude oil consumption by incorporating key determinants, including oil price, gross domestic product (GDP), inflation rates, population, geopolitical risk, and renewable energy usage. Multivariate Adaptive Regression Splines (MARS) approach is adopted in this study. MARS utilizes a piecewise-linear framework to effectively capture non-linear relationships among variables. The developed model consists of eight basis functions, including the intercept, and is structured as a first-order MARS model. The obtained results demonstrate that the model effectively captures the correlation between independent variables and crude oil consumption. Conducted sensitivity analysis on the independent variables revealed that population is the most influential factor, followed by oil price, while GDP and inflation rate exhibit moderate effects. Interestingly, renewable energy usage did not show significant impact on global crude oil consumption with the model, suggesting that crude oil will continue to play a dominant role in meeting future energy demand. The model’s predictive capability was validated using year 2025 data. The model prediction achieves high accuracy with an error margin of only 3.16%. Overall, the proposed method provides a simple, user-friendly correlation while offering valuable insights into the relative importance of key influencing factors in global crude oil consumption.
No takes yet. Share an insight, caveat, or question.
Malallah et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: