Deep learning-based electricity price forecasting: Findings on price predictability and European electricity markets

Kasra Aliyon and Jouni Ritvanen

Energy, 2024, vol. 308, issue C

Abstract: Deep learning algorithms have transformed day-ahead electricity price predictions in recent years. This phenomenon has implications for price predictability and perceived risks, which in turn could impact investments. This study aims to unravel the impact of deep learning-based price forecasting (DPF) on price predictability and its correlation with price volatility. This has been done by examining 19 bidding zones across 11 European countries from January 2018 to June 2023. This timeframe captures the unique market dynamics influenced by the COVID-19 pandemic and the Global Energy Crisis. This analysis has been enabled by the development of the Day-ahead European Electricity Price Forecasting Kit (Deepforkit), a forecasting tool developed in-house and open-sourced. Notably, Deepforkit offers a running speed at least 5 times faster than a previously suggested benchmark, enabling large-scale and multi-market analyses such as the one conducted in this study. It is quantitatively proven that if DPF is used, price volatility is just moderately correlated to unpredictability, and using it as a proxy for price unpredictability is not reliable. This is attributed to the capacity of DPFs to adjust their predictions in response to evolving market trends and patterns. This finding has an important implication on the investment risk perception and the cost of capital for new investments in the power sector. Moreover, the predictability behavior of prices is compared and contrasted in 3 markets, namely EPEX SPOT, OMIE, and Nord Pool, and across 11 European countries (Belgium, Denmark, Finland, France, Germany, Netherlands, Norway, Portugal, Sweden, Spain, and Switzerland).

Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544224026513
Full text for ScienceDirect subscribers only

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:eee:energy:v:308:y:2024:i:c:s0360544224026513

DOI: 10.1016/j.energy.2024.132877

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().