Benchmarking Time Series Foundation Models for Electricity Price Forecasting

Time series foundation models (TSFMs) have shown strong zero-shot forecasting performance, but their generalization in covariate-driven, non- stationary settings is underexplored. Electricity price forecasting (EPF) presents a challenging testbed due to complex temporal dependencies, distributional shifts, and strong reliance on struc- tural and contextual information. We propose a two-dataset-benchmarking framework for EPF to mitigate contamination risk and enable fair evalu- ation of TSFMs. We further examine key aspects of EPF including probabilistic forecasting perfor- mance, tail behavior, price spikes, and compar- isons against domain-specific methods. Overall, we find that TSFMs are competitive and often outperform general-purpose baselines, but their performance depends critically on covariate sup- port and they do not consistently surpass simpler, domain-specific methods tailored to EPF.