Probabilistic intraday electricity price forecasting using generative machine learning

2025-05-28

AI Summary

European intraday electricity markets require highly accurate price forecasts to support trading decisions under inherent uncertainty. Method: This paper introduces the first generative neural network model tailored to Germany’s continuous intraday electricity market, directly forecasting probabilistic price trajectories—rather than point estimates—by integrating generative modeling with trajectory-based probabilistic representation. We design a fixed-volume trading strategy and establish a dual-dimensional evaluation framework encompassing statistical metrics (e.g., Continuous Ranked Probability Score, CRPS) and economic performance (realized profit-and-loss from backtested trades). Contribution/Results: The model achieves statistical accuracy comparable to state-of-the-art benchmarks while delivering significantly higher trading profits in realistic market backtests. This demonstrates the tangible economic value of generative approaches for uncertainty quantification and decision support. By moving beyond conventional deterministic or ensemble-based forecasting paradigms, our work establishes a scalable, decision-oriented framework for intelligent electricity market operations.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

The growing importance of intraday electricity trading in Europe calls for improved price forecasting and tailored decision-support tools. In this paper, we propose a novel generative neural network model to generate probabilistic path forecasts for intraday electricity prices and use them to construct effective trading strategies for Germany's continuous-time intraday market. Our method demonstrates competitive performance in terms of statistical evaluation metrics compared to two state-of-the-art statistical benchmark approaches. To further assess its economic value, we consider a realistic fixed-volume trading scenario and propose various strategies for placing market sell orders based on the path forecasts. Among the different trading strategies, the price paths generated by our generative model lead to higher profit gains than the benchmark methods. Our findings highlight the potential of generative machine learning tools in electricity price forecasting and underscore the importance of economic evaluation.

Problem

• Improving probabilistic intraday electricity price forecasting
• Developing generative neural network for trading strategies
• Evaluating economic value of price path forecasts

Innovation

• Generative neural network for price forecasting
• Probabilistic path forecasts for trading strategies
• Economic evaluation of machine learning tools