Forecasting Day-Ahead Carbon Price by Modelling Its Determinants Using the PCA-Based Approach

Katarzyna Rudnik, Anna Hnydiuk-Stefan (), Aneta Kucińska-Landwójtowicz and Łukasz Mach
Additional contact information
Katarzyna Rudnik: Faculty of Production Engineering and Logistics, Opole University of Technology, 45-758 Opole, Poland
Anna Hnydiuk-Stefan: Faculty of Production Engineering and Logistics, Opole University of Technology, 45-758 Opole, Poland
Aneta Kucińska-Landwójtowicz: Faculty of Production Engineering and Logistics, Opole University of Technology, 45-758 Opole, Poland
Łukasz Mach: Faculty of Economics and Management, Opole University of Technology, 45-036 Opole, Poland

Energies, 2022, vol. 15, issue 21, 1-23

Abstract: Accurate price forecasts on the EU ETS market are of interest to many production and investment entities. This paper describes the day-ahead carbon price prediction based on a wide range of fuel and energy indicators traded on the Intercontinental Exchange market. The indicators are analyzed in seven groups for individual products (power, natural gas, coal, crude, heating oil, unleaded gasoline, gasoil). In the proposed approach, by combining the Principal Component Analysis (PCA) method and various methods of supervised machine learning, the possibilities of prediction in the period of rapid price increases are shown. The PCA method made it possible to reduce the number of variables from 37 to 4, which were inputs for predictive models. In the paper, these models are compared: regression trees, ensembles of regression trees, Gaussian Process Regression (GPR) models, Support Vector Machines (SVM) models and Neural Network Regression (NNR) models. The research showed that the Gaussian Process Regression model turned out to be the most advantageous and its price prediction can be considered very accurate.

Keywords: PCA; machine learning; time series forecasting; EU ETS; CO 2 emissions (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.mdpi.com/1996-1073/15/21/8057/pdf (application/pdf)
https://www.mdpi.com/1996-1073/15/21/8057/ (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:jeners:v:15:y:2022:i:21:p:8057-:d:957622

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

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