Oil Commodity Movement Estimation: Analysis with Gaussian Process and Data Science

Mulue Gebreslasie and Indranil SenGupta ()
Additional contact information
Mulue Gebreslasie: Department of Mathematics, North Dakota State University, Fargo, ND 58108, USA
Indranil SenGupta: Department of Mathematics and Statistics, Hunter College, City University of New York (CUNY), New York City, NY 10065, USA

Commodities, 2025, vol. 4, issue 2, 1-17

Abstract: In this study, Gaussian process (GP) regression is used to normalize observed commodity data and produce predictions at densely interpolated time intervals. The methodology is applied to an empirical oil price dataset. A Gaussian kernel with data-dependent initialization is used to calculate prediction means and confidence intervals. This approach generates synthetic data points from the denoised dataset to improve prediction accuracy. From this augmented larger dataset, a procedure is developed for estimating an upcoming crash-like behavior of the commodity price. Finally, multiple data-science-driven algorithms are used to demonstrate how data densification using GP regression improves the detection of forthcoming large fluctuations in a particular commodity dataset.

Keywords: crude oil; Gaussian process; synthetic data; crash indicator; data science (search for similar items in EconPapers)
JEL-codes: C0 C1 C2 C3 C4 C5 C6 C7 C8 C9 D4 E3 E6 F0 F1 F3 F4 F5 F6 G1 O1 O5 Q1 Q2 Q4 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2813-2432/4/2/9/pdf (application/pdf)
https://www.mdpi.com/2813-2432/4/2/9/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jcommo:v:4:y:2025:i:2:p:9-:d:1677378

Access Statistics for this article

Commodities is currently edited by Ms. Wendy Yu

More articles in Commodities from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().