Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Electricity Price Forecasting with Neural Networks on EPEX Order Books

: Schnürch, Simon; Wagner, Andreas

Volltext ()

Applied mathematical finance 27 (2020), Nr.3, S.189-206
ISSN: 1350-486X (Print)
ISSN: 1466-4313 (Online)
Bundesministerium für Bildung und Forschung BMBF (Deutschland)
Bundesministerium für Bildung und Forschung BMBF (Deutschland)
Zeitschriftenaufsatz, Elektronische Publikation
Fraunhofer ITWM ()

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 total demand. Appropriate feature extraction for the order book data is developed proceeding from existing literature. Using cross-validation to optimize hyperparameters, neural networks and random forests are fit to the data. Their in-sample and out-of-sample performance is compared to statistical reference models. The machine learning models outperform traditional approaches.