Regression Models Meet Foundation Models: A Hybrid-AI Approach to Practical Electricity Price Forecasting

2026-03-06

AI Summary

Electricity market prices exhibit high volatility, nonlinearity, and nonstationarity, posing significant challenges for traditional time series foundation models, which struggle to capture cross-variable dependencies and aperiodic patterns, while regression models are constrained by their reliance on only future-available features. To address these limitations, this work proposes FutureBoosting, a novel framework that synergistically integrates a frozen time series foundation model (TSFM) with a lightweight regression model. The TSFM encodes historical dependencies and generates predictive features that are injected as enriched inputs into the downstream regressor, substantially enhancing its temporal context awareness without requiring fine-tuning of the large model. Evaluated on real-world electricity market data, the proposed approach reduces mean absolute error (MAE) by over 30% compared to state-of-the-art baselines, demonstrating superior accuracy alongside efficiency, interpretability, and practical applicability.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Multiagent Systems: Other Foundations of Multi Agent Systems
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Electricity market prices exhibit extreme volatility, nonlinearity, and non-stationarity, making accurate forecasting a significant challenge. While cutting-edge time series foundation models (TSFMs) effectively capture temporal dependencies, they typically underutilize cross-variate correlations and non-periodic patterns that are essential for price forecasting. Conversely, regression models excel at capturing feature interactions but are limited to future-available inputs, ignoring crucial historical drivers that are unavailable at forecast time. To bridge this gap, we propose FutureBoosting, a novel paradigm that enhances regression-based forecasts by integrating forecasted features generated from a frozen TSFM. This approach leverages the TSFM's ability to model historical patterns and injects these insights as enriched inputs into a downstream regression model. We instantiate this paradigm into a lightweight, plug-and-play framework for electricity price forecasting. Extensive evaluations on real-world electricity market data demonstrate that our framework consistently outperforms state-of-the-art TSFMs and regression baselines, achieving reductions in Mean Absolute Error (MAE) of more than 30% at most. Through ablation studies and explainable AI (XAI) techniques, we validate the contribution of forecasted features and elucidate the model's decision-making process. FutureBoosting establishes a robust, interpretable, and effective solution for practical market participation, offering a general framework for enhancing regression models with temporal context.

Problem

• electricity price forecasting
• time series foundation models
• regression models
• non-stationarity
• feature interactions

Innovation

• FutureBoosting
• time series foundation models
• regression models
• electricity price forecasting
• hybrid AI