Foundation models promise accurate forecasts with little or no task-specific training, but whether they can replace models designed specifically for electricity price forecasting remains unclear. We compare nine variants from five foundation model families, evaluated in zero-shot mode, with two state-of-the-art electricity price forecasting benchmarks in Germany, Poland, and Spain over 2021-2025. Their performance is assessed in terms of point and probabilistic forecasting accuracy, as well as economic value in battery energy storage arbitrage. Only the TabPFN models consistently and significantly outperform the benchmarks across all three markets and all statistical measures. However, this statistical dominance does not translate directly into economic dominance: TabPFN performs best under unlimited bids and riskier quantile-based strategies, whereas the Distributional Deep Neural Network benchmark is more profitable when risk tolerance is lower. Thus, foundation models cannot universally replace market-specific models, and their value depends on both model architecture and the decision problem.
This study evaluates the forecasting performance of classical econometric, deep learning, convolutional, and Transformer-based models for aluminum futures prices to demonstrate that parsimonious forecasting approaches can remain effective competitors to substantially more complex machine learning architectures in alumi...
László Vancsura, Tibor Tatay, Tibor Bareith et al.· Decision Making Advances· 0 citations
Additional lag-only and lag-plus-calendar benchmarks show that price memory forms the predictive core of the problem, but that the full model still provides statistically significant incremental gains, especially in high-renewable, high-volatility, peak-hour, and upper-tail conditions.
Moein Jazayeri, Kian Jazayeri· International Journal of Eme...· 0 citations
A comparative evaluation of six deep learning models--covering state-space, MLP, RNN, and Transformer-based architectures--emphasizing generalization across markets suggests that N-HiTS and NBEATSx perform competitively in limited-data scenarios, while transformer-based models can reach comparable accuracy but tend to...
Hadeer El Ashhab, Sai Srijan Papineni, M. Dorn et al.· IEEE Access· 0 citations
It is suggested that the over-parameterization typically associated with regularization methods is not a necessary price for predictive accuracy in electricity price forecasting and that BMT offers researchers and practitioners a substantially more interpretable and computationally efficient alternative to shrinkage me...
It is concluded that investing in AI price forecasting software is highly viable, providing power traders, utilities, and grid operators with enhanced hedging capabilities and minimized imbalance charges.
Nehasri Thangirala, Komati Sridhar, K. Archana· American Journal of AI Cyber...· 0 citations
The growing penetration of variable renewable energy (VRE) is increasing the frequency of very low and negative prices, although these events also depend on demand, transmission capacity, price-regime persistence and flexibility resources. This study examines which pre-auction and diagnostic variables are associated wi...
T. Rokicki, P. Bórawski, Aneta Bełdycka-Bórawska et al.· Applied Sciences· 1 citation
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 23, 2026
Assistant Professor Benjamin Lindquist studies how engineers deliberately injected randomness into early AI, giving rise to deep-learning models that sometimes hallucinate.
