Parametric and Generative Forecasts of EPEX Dayhar45 Ahead Energy Market Curves

arXiv:2601.20226v2 Announce Type: replace Abstract: We propose two methodologies for modelling aggregated supply and demand curves in the EPEX SPOT Dayhar45 Ahead market, emphasizing generative models as a way to recover distributional variability. The first is a lowhar45 dimensional parametric representation that yields deterministic point forecasts; the second is a highhar45 dimensional orderhar45 level representation that samples from a conditional distribution of plausible curves. Both model the full curve structure, enabling the analysis of price sensitivity, volume sensitivity, and price impact. The parametric representation uses plateau levels, elastichar45 region boundaries, and polynomial coefficients, forecast with eXtreme Gradient Boosting. The main contribution is the generative representation, which uses price arrivals and volumehar45 increment marks and is implemented with conditional Denoising Diffusion Probabilistic Models. Using French EPEX data from 2021 to 2024, we evaluate both approaches through curve reconstruction and a pricehar45 maker storage optimization problem. The parametric implementation provides a deterministic reference, while the diffusionhar45 based implementation produces distributions of plausible curves and achieves higher realized profits and smaller gaps to an oracle benchmark in the storage application.

Source: arXiv cs.LG | 2026-07-09