Variance Stabilizing Transformations for Electricity Price Forecasting in Periods of Increased Volatility

2025-11-17

AI Summary

High volatility in electricity prices—driven by high renewable penetration and market disturbances—severely degrades the accuracy of existing forecasting models. To address this, we propose a parametric inverse hyperbolic sine (asinh) variance-stabilizing transformation (VST), systematically analyzing parameter sensitivity and calibration window effects, and augmenting it with a rolling-average-based multi-transformation fusion strategy to enhance model robustness. We empirically evaluate the approach on day-ahead electricity price data from Germany, Spain, and France (2015–2024), integrated with LEAR and NARX models. Results show that, relative to the untransformed baseline, the proposed method reduces LEAR prediction error by up to 14.6% and NARX error by up to 8.7%. The synergistic combination of parametric asinh VST and rolling-average fusion achieves up to 17.7% error reduction, markedly improving stability during periods of extreme price volatility. This work establishes an interpretable, easily calibratable preprocessing paradigm for forecasting under high uncertainty in modern power markets.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

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

Abstract

Accurate day-ahead electricity price forecasts are critical for power system operation and market participation, yet growing renewable penetration and recent crises have caused unprecedented volatility that challenges standard models. This paper revisits variance-stabilizing transformations (VSTs) as a preprocessing tool by introducing a novel parametrization of the asinh transformation, systematically analyzing parameter sensitivity and calibration window size, and explicitly testing performance under volatile market regimes. Using data from Germany, Spain, and France over 2015-2024 with two model classes (NARX and LEAR), we show that VSTs substantially reduce forecast errors, with gains of up to 14.6% for LEAR and 8.7% for NARX relative to untransformed benchmarks. The new parametrized asinh consistently outperforms its standard form, while rolling averaging across transformations delivers the most robust improvements, reducing errors by up to 17.7%. Results demonstrate that VSTs are especially valuable in volatile regimes, making them a powerful tool for enhancing electricity price forecasting in today's power markets.

Problem

• Addressing electricity price volatility from renewables and crises
• Improving forecasting accuracy with variance-stabilizing transformations
• Enhancing model robustness during high market volatility periods

Innovation

• Introduces novel parametrization of asinh transformation
• Systematically analyzes parameter sensitivity and calibration window
• Uses rolling averaging across transformations for robust improvements