IMMORTAL - Untersuchung der Marktmikrostruktur und Kurzfrist Preisvorhersagen für Intra-day Elektrizitätsmärkte

• Antragsteller: Professor Dr. Rüdiger Kiesel; Professor Dr. Florian Ziel
• Fachliche Zuordnung: Statistik und Ökonometrie
• Förderung: Förderung von 2017 bis 2022
• Projektkennung: Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 379008354

• Erstellungsjahr: 2023

Zusammenfassung der Projektergebnisse

The main aim of the project was to adequately address two pertinent challenges: the enhancement of the understanding of the intraday electricity market microstructure as well as the development of innovative forecasting methods that meet the very specific characteristics of intraday electricity trading. The trading intensity on the intraday electricity market can be modeled by point processes. We extended the frequently used self-exciting point process model, which accounts for the clustering of orders that is observed on the market, by additional exogenous factors. For this purpose, we considered fundamental as well as market-induced factors, such as forecasting errors in renewable power generation or volumes activated on the balancing market. On the one hand, we found that adding a self-exciting term to models that solely take into account exogenous impacts improves the models. On the other hand, purely self-exciting models need to be enhanced by additional exogenous factors to capture the market dynamics correctly. Furthermore, our results imply that if the trading intensity is linked to an unobservable stochastic factor process, the usage of gradient-based optimization algorithms combined with automatic differentiation to obtain exact derivatives is beneficial compared to derivative-free approaches used for model calibration. In addition, intraday electricity price forecasting models were designed. They take into account autoregressive and calendar based deterministic effects as well as relevant external inputs. The latter include mainly wind, solar, and demand forecasts such as power plant outages. The empirical models show that all components are relevant especially in driving the market volatility. Next to pure forecasting models we also evaluated the economic benefit of a better forecast accuracy in intraday electricity markets. We contributed to better short-term electricity trading strategies for risk neutral and risk averse market participants. The results show improvement over state-of-the-art trading strategies, especially for large market participants. Finally, we want to remark that several empirical studies of the German and Polish IMMORTAL research results showed that the intraday electricity market is (almost) efficient which is a valuable information for market operators and regulators.

Projektbezogene Publikationen (Auswahl)

• The value of forecasts: Quantifying the economic gains of accurate quarter-hourly electricity price forecasts. Energy Economics, 76, 411-423.
  Kath, Christopher & Ziel, Florian
  
• Electricity price forecasting. Routledge Handbook of Energy Economics, 506-521. Routledge.
  Weron, Rafał & Ziel, Florian
  
• Estimation and Simulation of the Transaction Arrival Process in Intraday Electricity Markets. Energies, 12(23), 4518.
  Narajewski, Michał & Ziel, Florian
  
• Econometric modelling and forecasting of intraday electricity prices. Journal of Commodity Markets, 19, 100107.
  Narajewski, Michał & Ziel, Florian
  
• Ensemble forecasting for intraday electricity prices: Simulating trajectories. Applied Energy, 279, 115801.
  Narajewski, Michał & Ziel, Florian
  
• Load Nowcasting: Predicting Actuals with Limited Data. Energies, 13(6), 1443.
  Ziel, Florian
  
• “Changes in electricity demand pattern in Europe due to COVID-19 shutdowns.” In: IAEE Energy Forum / Covid-19 Issue 2020
  Narajewski, M. & Ziel, F.
  
• “Optimal order execution in intraday markets: Minimizing costs in trade trajectories.” (2020)
  Kath, C. & Ziel, F.
  
• Exogenous factors for order arrivals on the intraday electricity market. Energy Economics, 97, 105186.
  Kramer, Anke & Kiesel, Rüdiger
  
• The Impact of Renewable Energy Forecasts on Intraday Electricity Prices. Economics of Energy & Environmental Policy, 10(1).
  Kulakov, Sergei & Ziel, Florian
  
• tsrobprep — an R package for robust preprocessing of time series data. SoftwareX, 16, 100809.
  Narajewski, Michał; Kley-Holsteg, Jens & Ziel, Florian
  
• Integrated day-ahead and intraday self-schedule bidding for energy storage systems using approximate dynamic programming. European Journal of Operational Research, 301(2), 726-746.
  Finnah, Benedikt; Gönsch, Jochen & Ziel, Florian
  
• Optimal bidding in hourly and quarter-hourly electricity price auctions: Trading large volumes of power with market impact and transaction costs. Energy Economics, 110, 105974.
  Narajewski, Michał & Ziel, Florian
  
• “Random intensity models with an application to intraday electricity markets.” Doctoral thesis. Universität Duisburg-Essen.
  Kramer, A.
  
• “Predictive Analytics in Short-term Electricity Markets.” Doctoral thesis. Universität Duisburg-Essen (2023)
  Narajewski, M.