Zero-shot time-series foundation models against the Khepri v1.5 baselines for 96-hour carbon-intensity forecasts in nine Nordic bidding zones: a preregistered comparison
Abstract
We compare three zero-shot time-series foundation models (Timestral-5M, Timestral-209M and Chronos-2, 8192-hour context, no covariates) with the Khepri v1.5 SARIMA and GBM baselines, with their code and hyperparameters unchanged, for 96-hour forecasts of hourly carbon intensity in the Nordic bidding zones NO1-NO5 and SE1-SE4. The design, the analysis code and a minimum detectable effect (MDE) were timestamped before any foundation-model forecast by an OpenTimestamps proof (Bitcoin-confirmed) and the GPG-signed tag prereg-v1; the proofs are archived in Zenodo (DOI 10.5281/zenodo.23881315). The preregistration was also submitted to OSF Registries 2026-11-02 07:49 UTC (osf.io/hncz5; embargo requested, pending OSF approval). On the 2025 test (361 daily origins, contamination not ruled out) all three models have a lower zone-averaged MAPE than both baselines, by 0.952 to 1.347 percentage points (Diebold-Mariano with the Harvey-Leybourne-Newbold correction, all p < 0.0001), and each beats SARIMA in 9 of 9 zones. On a post-release holdout of 14 origins from 2026-10-16 the point estimates have the same sign (0.729 to 1.326 percentage points lower), but all lie within the preregistered MDE of 2.989 points and are reported as not determined. For calibration (RQ4) the preregistration locked a Diebold-Mariano-HLN test on CRPS; the MDE applies to MAPE only. Holdout CRPS point estimates (n = 14) are 1.1114 (Timestral-5M), 1.0733 (Timestral-209M) and 1.1198 (Chronos-2) against 1.4165 gCO2-eq/kWh for SARIMA (each p < 0.0001). The SARIMA quantiles are constructed from the fitted model's Gaussian predictive distribution (deviation D-03), not taken from a native v1.5 forecast, so the difference may partly reflect the SARIMA intervals. The contamination indicator does not exceed its preregistered threshold for any comparison; with 14 holdout origins it has low power (HAC standard errors 0.124 to 0.360 pp).
Related
- Cosmological Inference with Gravitational-Wave Dark Sirens and 21-cm Intensity Mapping: Forecasts for LIGO-Virgo-KAGRA with MeerKAT and SKAO
- Multivariate phase-space interpretation of ensemble forecasts: A comparison of physical and AI models
- Dosimetric Comparison of Intensity-Modulated Radiotherapy and Four-Field Three-Dimensional Conformal Radiotherapy in the Management of Prostate Cancer: A Retrospective Dual-Centre Analysis
- In Silico Dosimetric Comparison of Intensity-Modulated Photon Versus Proton Therapy for Pelvic and Para-Aortic Nodal Recurrence of Prostate Cancer Following Prior Irradiation
- Gaussian process regression with physics-guided features for WLAM bead geometry: Repeated validation, algorithmic baselines, and predictive-interval assessment
BibTeX
@misc{nicolaysen2026zeroshot,
title = {{Zero-shot time-series foundation models against the Khepri v1.5 baselines for 96-hour carbon-intensity forecasts in nine Nordic bidding zones: a preregistered comparison}},
author = {Eirik Botten Nicolaysen},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.02895},
eprint = {2611.02895},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.02895},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Zero-shot time-series foundation models against the Khepri v1.5 baselines for 96-hour carbon-intensity forecasts in nine Nordic bidding zones: a preregistered comparison
AU - Eirik Botten Nicolaysen
PY - 2026
DA - 2026/11/02
PB - arXiv
DO - 10.48550/arxiv.2611.02895
UR - https://arxiv.org/abs/2611.02895
AB - We compare three zero-shot time-series foundation models (Timestral-5M, Timestral-209M and Chronos-2, 8192-hour context, no covariates) with the Khepri v1.5 SARIMA and GBM baselines, with their code and hyperparameters unchanged, for 96-hour forecasts of hourly carbon intensity in the Nordic bidding zones NO1-NO5 and SE1-SE4. The design, the analysis code and a minimum detectable effect (MDE) were timestamped before any foundation-model forecast by an OpenTimestamps proof (Bitcoin-confirmed) and the GPG-signed tag prereg-v1; the proofs are archived in Zenodo (DOI 10.5281/zenodo.23881315). The preregistration was also submitted to OSF Registries 2026-11-02 07:49 UTC (osf.io/hncz5; embargo requested, pending OSF approval). On the 2025 test (361 daily origins, contamination not ruled out) all three models have a lower zone-averaged MAPE than both baselines, by 0.952 to 1.347 percentage points (Diebold-Mariano with the Harvey-Leybourne-Newbold correction, all p < 0.0001), and each beats SARIMA in 9 of 9 zones. On a post-release holdout of 14 origins from 2026-10-16 the point estimates have the same sign (0.729 to 1.326 percentage points lower), but all lie within the preregistered MDE of 2.989 points and are reported as not determined. For calibration (RQ4) the preregistration locked a Diebold-Mariano-HLN test on CRPS; the MDE applies to MAPE only. Holdout CRPS point estimates (n = 14) are 1.1114 (Timestral-5M), 1.0733 (Timestral-209M) and 1.1198 (Chronos-2) against 1.4165 gCO2-eq/kWh for SARIMA (each p < 0.0001). The SARIMA quantiles are constructed from the fitted model's Gaussian predictive distribution (deviation D-03), not taken from a native v1.5 forecast, so the difference may partly reflect the SARIMA intervals. The contamination indicator does not exceed its preregistered threshold for any comparison; with 14 holdout origins it has low power (HAC standard errors 0.124 to 0.360 pp).
ER -