The Evolving Clean Energy Landscape: Drivers, Trends, and Future Prospects

Abstract

The rapid growth of the clean energy sector raises critical questions about the factors shaping its financial performance. Prior studies have documented that clean energy stock returns are primarily driven by market uncertainty and financial distress, particularly in the context of small-cap stocks and emerging markets, rather than by conventional determinants such as oil prices (Ciner et al., 2023). Building on this, the first study examines a broad set of financial drivers, including macroeconomic conditions, policy uncertainty, technological innovation, market volatility, global economic trends, oil prices, green stocks, and emerging market equities. Additionally, it incorporates insights from the GDELT database, capturing sentiment indicators from media reports and biases in political coverage. To preliminarily validate these factors, this study applies unsupervised learning techniques and employs Extreme Bound Analysis (EBA) to assess whether both established and novel factors significantly impact the clean energy market. This approach extends prior research and provides a foundation for future studies. The findings on traditional and emerging factors further inform clean energy market forecasting. Given the widely use of machine learning, our second study integrates statistical, machine learning, and hybrid models to identify the most effective forecasting approach. Model performance is evaluated based on forecasting horizons, accuracy, and trend estimation differences. This comprehensive assessment aims to determine the most suitable model for predicting the clean energy market under various conditions. The first study has primarily analyzed the relationship between media reporting preferences, sentiment values, and the clean energy market. In the third stage of research, the scope of data collection will be expanded to include models that measure the strength of policy commitments, introducing a novel perspective on the clean energy market from a political science standpoint. Specifically, our third study intends to examine textual reports of speeches by central bankers and employ advanced language models to assess key factors such as urgency (how soon), intensity (how much), and timeliness (how long). By incorporating this policy commitment dimension, the research aims to identify additional driving forces influencing the clean energy market. This thesis advances understanding of the clean energy market by combining traditional and emerging drivers with insights from media coverage and political discourse, addressing a dimension that has received limited scholarly attention. It shows how positive narratives and credible policy commitments enhance transparency, attract responsible investors, and shape portfolio decisions, creating a self-reinforcing cycle that accelerates the clean energy transition. Using machine learning and statistical models, it also provides a methodological basis for forecasting market volatility, offering practical insights for investors, policymakers, and researchers.