Cargando...
Fecha
2026-02-01
Derechos de acceso
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editorial
Citas
Resumen
Este trabajo analiza la evolución del riesgo sistémico en los mercados energéticos europeos en un contexto marcado por episodios recientes de elevada incertidumbre geopolítica y económica. En particular, se estudia el impacto de tres eventos de gran alcance —el inicio de la pandemia de COVID-19, la invasión de Ucrania por parte de Rusia y el cambio en la política arancelaria de Estados Unidos en 2025— sobre la dinámica y los mecanismos de contagio de riesgo entre distintas variables representativas de los mercados de combustibles fósiles, energías limpias y factores de riesgo financiero.
Con este objetivo, se propone un enfoque metodológico basado en regresión cuantílica no lineal mediante redes neuronales recurrentes de tipo xLSTM, que permite estimar de forma dinámica el Value at Risk (VaR) y el Conditional Value at Risk (CoVaR) de las variables anteriores. A partir de las derivadas parciales del modelo entrenado se construye una red de contagio sistémico, de la que se derivan tres indicadores de riesgo: el Systemic Fragility Index (SFI), el Systemic Hazard Index (SHI) y el Systemic Network Risk Index (SNRI).
Los resultados muestran que la invasión de Ucrania constituye el shock más severo desde el punto de vista sistémico, con un aumento significativo del riesgo agregado y una intensificación de las interdependencias entre variables, especialmente en el precio del gas natural. En contraste, la crisis asociada a la COVID-19 presenta un comportamiento transitorio, con una reducción posterior del riesgo sistémico agregado. Asimismo, se observa que, de los factores estudiados, las principales aportaciones al riesgo geopolítico provienen del precio del crudo y el tipo cambiario dólar estadounidense-rublo.
El trabajo contribuye a la literatura proporcionando un marco flexible para el análisis del riesgo sistémico en entornos complejos y ofrece implicaciones relevantes para la gestión del riesgo y el diseño de políticas energéticas en la Unión Europea.
This paper analyzes the evolution of systemic risk in European energy markets in a context shaped by recent episodes of heightened geopolitical and economic uncertainty. In particular, it examines the impact of three major events—the onset of the COVID-19 pandemic, Russia’s invasion of Ukraine, and the change in U.S. tariff policy in 2025—on the dynamics and mechanisms of risk contagion among a set of variables representative of fossil fuel markets, clean energy markets, and risk factors. To this end, the paper proposes a methodological framework based on nonlinear quantile regression implemented through xLSTM recurrent neural networks, which enables the dynamic estimation of the Value at Risk (VaR) and the Conditional Value at Risk (CoVaR) of the aforementioned variables. Using the partial derivatives of the trained model, a systemic contagion network is constructed, from which three risk indicators are derived: the Systemic Fragility Index (SFI), the Systemic Hazard Index (SHI), and the Systemic Network Risk Index (SNRI). The results show that the invasion of Ukraine constitutes the most severe shock from a systemic perspective, leading to a significant increase in aggregate risk and a strengthening of interdependencies across variables, particularly in natural gas prices. In contrast, the crisis associated with COVID-19 exhibits a transitory pattern, followed by a subsequent reduction in aggregate systemic risk. Furthermore, among the factors analyzed, the main contributions to geopolitical risk stem from crude oil prices and the U.S. dollar–Russian ruble exchange rate. This study contributes to the literature by providing a flexible framework for the analysis of systemic risk in complex environments and offers relevant implications for risk management and the design of energy policies in the European Union.
This paper analyzes the evolution of systemic risk in European energy markets in a context shaped by recent episodes of heightened geopolitical and economic uncertainty. In particular, it examines the impact of three major events—the onset of the COVID-19 pandemic, Russia’s invasion of Ukraine, and the change in U.S. tariff policy in 2025—on the dynamics and mechanisms of risk contagion among a set of variables representative of fossil fuel markets, clean energy markets, and risk factors. To this end, the paper proposes a methodological framework based on nonlinear quantile regression implemented through xLSTM recurrent neural networks, which enables the dynamic estimation of the Value at Risk (VaR) and the Conditional Value at Risk (CoVaR) of the aforementioned variables. Using the partial derivatives of the trained model, a systemic contagion network is constructed, from which three risk indicators are derived: the Systemic Fragility Index (SFI), the Systemic Hazard Index (SHI), and the Systemic Network Risk Index (SNRI). The results show that the invasion of Ukraine constitutes the most severe shock from a systemic perspective, leading to a significant increase in aggregate risk and a strengthening of interdependencies across variables, particularly in natural gas prices. In contrast, the crisis associated with COVID-19 exhibits a transitory pattern, followed by a subsequent reduction in aggregate systemic risk. Furthermore, among the factors analyzed, the main contributions to geopolitical risk stem from crude oil prices and the U.S. dollar–Russian ruble exchange rate. This study contributes to the literature by providing a flexible framework for the analysis of systemic risk in complex environments and offers relevant implications for risk management and the design of energy policies in the European Union.
Descripción
Categorías UNESCO
Palabras clave
Contagio riesgo sistémico, CoVaR, Regresión cuantílica, LSTM, Mercado energético europeo, Systemic risk spillover, CoVaR, Quantile regression, LSTM, European energy markets
Citación
Lozano Fernández, Fernando. Trabajo de Fin de Máster: Evaluación de la exposición al riesgo en los mercados energéticos europeos. Universidad Nacional de Educación a Distancia (UNED), 2026
Centro
Escuela Técnica Superior de Ingeniería Informática
Departamento
Inteligencia Artificial