ReModels: Quantile Regression Averaging models

Abstract

Electricity price forecasts are essential for making informed business decisions within the electricity markets. Probabilistic forecasts, which provide a range of possible future prices rather than a single estimate, are particularly valuable for capturing market uncertainties. The Quantile Regression Averaging (QRA) method is a leading approach to generating these probabilistic forecasts. In this paper, we introduce ReModels, a comprehensive Python package that implements QRA and its various modifications from recent literature. This package not only offers tools for QRA but also includes features for data acquisition, preparation, and variance stabilizing transformations (VSTs). To the best of our knowledge, there is no publicly available implementation of QRA and its variants. Our package aims to fill this gap, providing researchers and practitioners with the tools to generate accurate and reliable probabilistic forecasts in the field of electricity price forecasting.

Keywords

• Machine learning
• Energy price forecasting
• Probabilistic forecasting
• Quantile regression
• Quantile regression averaging
• QRA