Application of LSTM Models in Forecasting Brent Crude Oil Prices During the Green Energy Transition
Year: 2024
Type: article
Abstract: Crude oil, as a vital strategic resource in the global energy and environmental landscape, significantly influences the global economy, financial markets, and progress toward low-carbon development. In recent years, the crude oil market has become increasingly complex due to the accelerating transition to green energy and low-carbon policies worldwide. Crude oil prices are shaped by traditional factors such as supply-demand dynamics, geopolitical tensions, and technological advancements, as well as emerging challenges like shifts in energy policies, carbon emissions regulations, and innovations in renewable energy technologies. This multifaceted interplay of drivers underscores the critical importance of crude oil price forecasting for economic planning, energy policy, and financial market stability. This study employs spot price data for European Brent crude oil, sourced from the U.S. Energy Information Administration, to develop a deep learning model comprising three LSTM layers for short-term price prediction. The findings demonstrate that the LSTM model effectively captures overall pricing trends, though it exhibits minor discrepancies during periods of significant price volatility. This research not only validates the LSTM model's utility in forecasting within energy markets but also offers valuable insights and data to support policymakers and investors navigating the uncertainties of crude oil price fluctuations.
Authors: Yuwen Zhao, Baojun Hu, Sizhe Wang
Institutions: The University of Sydney, University of Florida, Lehigh University
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4405591088