Assessing Electricity Demand Forecasting with Exogenous Data in Time Series Foundation Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910012722380800 |
|---|---|
| author | Cheong, Wei Soon Jiang, Lian Lian Ling, Jamie Ng Suat |
| author_facet | Cheong, Wei Soon Jiang, Lian Lian Ling, Jamie Ng Suat |
| contents | Time-series foundation models have emerged as a new paradigm for forecasting, yet their ability to effectively leverage exogenous features -- critical for electricity demand forecasting -- remains unclear. This paper empirically evaluates foundation models capable of modeling cross-channel correlations against a baseline LSTM with reversible instance normalization across Singaporean and Australian electricity markets at hourly and daily granularities. We systematically assess MOIRAI, MOMENT, TinyTimeMixers, ChronosX, and Chronos-2 under three feature configurations: all features, selected features, and target-only. Our findings reveal highly variable effectiveness: while Chronos-2 achieves the best performance among foundation models (in zero-shot settings), the simple baseline frequently outperforms all foundation models in Singapore's stable climate, particularly for short-term horizons. Model architecture proves critical, with synergistic architectural implementations (TTM's channel-mixing, Chronos-2's grouped attention) consistently leveraging exogenous features, while other approaches show inconsistent benefits. Geographic context emerges as equally important, with foundation models demonstrating advantages primarily in variable climates. These results challenge assumptions about universal foundation model superiority and highlight the need for domain-specific models, specifically in the energy domain. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_05390 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Assessing Electricity Demand Forecasting with Exogenous Data in Time Series Foundation Models Cheong, Wei Soon Jiang, Lian Lian Ling, Jamie Ng Suat Machine Learning Artificial Intelligence G.3; I.2.1 Time-series foundation models have emerged as a new paradigm for forecasting, yet their ability to effectively leverage exogenous features -- critical for electricity demand forecasting -- remains unclear. This paper empirically evaluates foundation models capable of modeling cross-channel correlations against a baseline LSTM with reversible instance normalization across Singaporean and Australian electricity markets at hourly and daily granularities. We systematically assess MOIRAI, MOMENT, TinyTimeMixers, ChronosX, and Chronos-2 under three feature configurations: all features, selected features, and target-only. Our findings reveal highly variable effectiveness: while Chronos-2 achieves the best performance among foundation models (in zero-shot settings), the simple baseline frequently outperforms all foundation models in Singapore's stable climate, particularly for short-term horizons. Model architecture proves critical, with synergistic architectural implementations (TTM's channel-mixing, Chronos-2's grouped attention) consistently leveraging exogenous features, while other approaches show inconsistent benefits. Geographic context emerges as equally important, with foundation models demonstrating advantages primarily in variable climates. These results challenge assumptions about universal foundation model superiority and highlight the need for domain-specific models, specifically in the energy domain. |
| title | Assessing Electricity Demand Forecasting with Exogenous Data in Time Series Foundation Models |
| topic | Machine Learning Artificial Intelligence G.3; I.2.1 |
| url | https://arxiv.org/abs/2602.05390 |