Machine Learning on EPEX Order Books: Insights and Forecasts

TL;DR

This paper develops machine learning models, including neural networks and random forests, using order book and fundamental data to improve forecasts of German electricity spot market prices, outperforming traditional statistical models.

Contribution

It introduces feature extraction methods for order book data and compares advanced machine learning models to statistical benchmarks for electricity price forecasting.

Findings

ML models outperform traditional statistical models

Neural networks and random forests show superior accuracy

Order book features enhance forecast quality

Abstract

This paper employs machine learning algorithms to forecast German electricity spot market prices. The forecasts utilize in particular bid and ask order book data from the spot market but also fundamental market data like renewable infeed and expected demand. Appropriate feature extraction for the order book data is developed. Using cross-validation to optimise hyperparameters, neural networks and random forests are proposed and compared to statistical reference models. The machine learning models outperform traditional approaches.