Empirical evaluation of Time Series Foundation Models for Day-ahead and Imbalance Electricity Price Forecasting in Belgium

TL;DR

This paper systematically evaluates Time Series Foundation Models for Belgian electricity price forecasting, highlighting their strengths and limitations in volatile market conditions.

Contribution

It provides the first comprehensive empirical assessment of TSFMs like Chronos-2, Chronos-Bolt, and TimesFM 2.5 in complex electricity markets.

Findings

Chronos-2 in ARX mode achieves the lowest MAE for day-ahead prices.

TSFMs outperform other models in zero-shot forecasting but struggle in extreme conditions.

Chronos-2's MAE is 5% lower for day-ahead and 10% higher for imbalance prices compared to ensemble methods.

Abstract

Recent advances in Time Series Foundation Models (TSFMs) promise zero-shot forecasting capabilities with minimal task-specific training. While these models have shown strong performance across generic benchmarks, their applicability in volatile, complex electricity markets remains underexplored. Addressing this gap, this study provides a systematic empirical evaluation of several TSFMs, specifically Chronos-2 and Chronos-Bolt (developed by Amazon), and TimesFM 2.5 (provided by Google), for forecasting Belgian day-ahead and imbalance electricity prices. For both considered markets, Chronos-2 in ARX mode produces the most accurate forecasts. Compared with the best ensemble prediction from other machine learning methods, Chronos-2's Mean Absolute Error (MAE) is 5% lower for the day-ahead market. In contrast, the model yields 10% higher MAE predicting imbalance prices across all forecast horizons, except for the two-hour-ahead horizon. Moreover, we find that TSFMs exhibit genuine zero-shot forecasting skills but still struggle under extreme market conditions.