 Forecasting crude oil prices in the COVID-19 era: Can machine learn better?Guangning Tian, Yuchao Peng and Yuhao Meng Energy Economics, 2023, vol. 125, issue C Abstract: Since the onset of the COVID-19 pandemic, energy price predictability has worsened. We evaluate the effectiveness of the two machine learning methods of shrinkage and combination on the spot prices of crude oil before and during the COVID-19 epidemic. The results demonstrated that COVID-19 increased economic uncertainty and diminished the predictive capacity of numerous models. Shrinkage methods have always been regarded as having an excellent out-of-sample forecast performance. However, during the COVID period, the combination methods provide more accurate information than the shrinkage methods. The reason is that the outbreak of the epidemic has altered the correlation between specific predictors and crude oil prices, and shrinkage methods are incapable of identifying this change, resulting in the loss of information. Keywords: Extreme events; Oil price forecast; Out-of-sample forecasts; Shrinkage methods; Combination methods (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:eee:eneeco:v:125:y:2023:i:c:s0140988323002864 DOI: 10.1016/j.eneco.2023.106788 Access Statistics for this article Energy Economics is currently edited by R. S. J. Tol, Beng Ang, Lance Bachmeier, Perry Sadorsky, Ugur Soytas and J. P. Weyant More articles in Energy Economics from Elsevier |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.