Price Forecasting in the German Electricity Market: The Impacts of the Energy Transition and Climate Change
(2024) DABN01 20241Department of Economics
Department of Statistics
- Abstract
- Modern societies rely on a stable supply of electric power from highly complex energy systems. Balancing demand and supply at electricity markets becomes increasingly challenging with the rising share of highly volatile renewable energy sources (RES) and the influence of climate change. Within this uncertain environment, actors face major challenges for planning and decision making. Thus, this thesis explores the price forecasting in the German electricity market, focusing on the effects of the energy transition and climate change. We employ several machine learning models, namely elastic net linear regression, XGBoost, artificial neural network, and generative adversarial network, to predict hourly day-ahead electricity prices. Our... (More)
- Modern societies rely on a stable supply of electric power from highly complex energy systems. Balancing demand and supply at electricity markets becomes increasingly challenging with the rising share of highly volatile renewable energy sources (RES) and the influence of climate change. Within this uncertain environment, actors face major challenges for planning and decision making. Thus, this thesis explores the price forecasting in the German electricity market, focusing on the effects of the energy transition and climate change. We employ several machine learning models, namely elastic net linear regression, XGBoost, artificial neural network, and generative adversarial network, to predict hourly day-ahead electricity prices. Our results show that XGBoost provides the most accurate forecasts among the models tested. The developed EPF-sWGAN model showed undesireable behavior and did not outperform traditional approaches. Our analysis reveals that increased RES generation lowers prices but increases price volatility. This necessitates enhanced risk management strategies for market participants. Additionally, factors like CO2 certificate prices, gas prices, and ambient temperature were found to have a lower but still notable influence on electricity prices. The findings underscore the potential of Germany's energy transition to decrease prices as well as the energy system's vulnerability to external factors, such as fluctuating gas prices driven by geopolitical issues. Future research directions include incorporating additional variables like electricity imports, exports, and storage capacities, as well as leveraging sentiment analysis to better capture market expectations and improve forecasting accuracy. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9160893
- author
- Behling, Magnus LU and Wagner, Christoph Gerhard Volker LU
- supervisor
-
- Simon Reese LU
- organization
- course
- DABN01 20241
- year
- 2024
- type
- H1 - Master's Degree (One Year)
- subject
- keywords
- Price Forecasting, German Electricity Market, Generative Adversarial Network
- language
- English
- id
- 9160893
- date added to LUP
- 2024-09-24 08:32:28
- date last changed
- 2024-09-24 08:32:28
@misc{9160893,
abstract = {{Modern societies rely on a stable supply of electric power from highly complex energy systems. Balancing demand and supply at electricity markets becomes increasingly challenging with the rising share of highly volatile renewable energy sources (RES) and the influence of climate change. Within this uncertain environment, actors face major challenges for planning and decision making. Thus, this thesis explores the price forecasting in the German electricity market, focusing on the effects of the energy transition and climate change. We employ several machine learning models, namely elastic net linear regression, XGBoost, artificial neural network, and generative adversarial network, to predict hourly day-ahead electricity prices. Our results show that XGBoost provides the most accurate forecasts among the models tested. The developed EPF-sWGAN model showed undesireable behavior and did not outperform traditional approaches. Our analysis reveals that increased RES generation lowers prices but increases price volatility. This necessitates enhanced risk management strategies for market participants. Additionally, factors like CO2 certificate prices, gas prices, and ambient temperature were found to have a lower but still notable influence on electricity prices. The findings underscore the potential of Germany's energy transition to decrease prices as well as the energy system's vulnerability to external factors, such as fluctuating gas prices driven by geopolitical issues. Future research directions include incorporating additional variables like electricity imports, exports, and storage capacities, as well as leveraging sentiment analysis to better capture market expectations and improve forecasting accuracy.}},
author = {{Behling, Magnus and Wagner, Christoph Gerhard Volker}},
language = {{eng}},
note = {{Student Paper}},
title = {{Price Forecasting in the German Electricity Market: The Impacts of the Energy Transition and Climate Change}},
year = {{2024}},
}