On-line conformalized neural networks ensembles for probabilistic forecasting of day-ahead electricity prices
Alessandro Brusaferri, Andrea Ballarino, Luigi Grossi, Fabrizio Laurini
TL;DR
Probabilistic forecasting of day-ahead electricity prices requires reliable uncertainty quantification under volatile market dynamics. The authors propose online conformalized neural network ensembles (OCQ) that integrate asymmetric conformalized quantile regression (CQR), online recalibration, non-exchangeable CP, and uniform vincentization to produce calibrated, horizon-conditioned prediction intervals with coverage $1-\alpha$. Across German GE and Italian bidding zones, OCQ methods achieve improved hourly coverage and stable probabilistic scores relative to QRA, Deep Quantile Regression, Distributional NNs, and standard CP variants, using a rolling online recalibration and ensemble averaging. The framework provides a scalable, distribution-free calibration remedy for PEPF under distribution shifts, with open-source code to support replication and extension.
Abstract
Probabilistic electricity price forecasting (PEPF) is subject of increasing interest, following the demand for proper quantification of prediction uncertainty, to support the operation in complex power markets with increasing share of renewable generation. Distributional neural networks ensembles have been recently shown to outperform state of the art PEPF benchmarks. Still, they require critical reliability enhancements, as fail to pass the coverage tests at various steps on the prediction horizon. In this work, we propose a novel approach to PEPF, extending the state of the art neural networks ensembles based methods through conformal inference based techniques, deployed within an on-line recalibration procedure. Experiments have been conducted on multiple market regions, achieving day-ahead forecasts with improved hourly coverage and stable probabilistic scores.