Welcome to the UPF Digital Repository

Modelling financial tail risk: multi-step forecasting of value-at-risk and expected shortfall with the realized Har-Garch model

Show simple item record

dc.contributor.author Penfold, Kai
dc.contributor.author Ravinet, Raphael
dc.date.accessioned 2024-08-27T11:53:30Z
dc.date.available 2024-08-27T11:53:30Z
dc.date.issued 2024-06
dc.identifier.uri http://hdl.handle.net/10230/60931
dc.description Treball fi de màster de: Master's Degree in Economics and Finance. Finance Programme. Curs 2023-2024
dc.description Director: Christian Brownlees
dc.description.abstract It has been well documented in the literature that the inclusion of realized measures into GARCH models can lead to both statistical and economic gains through improved forecasting performance. In this paper, we employ the Realized HAR-GARCH model of Huang et al. (2016) to forecast multi-step-ahead financial tail risk. Specifically, we assess the model’s performance and whether the resulting Value-at-Risk (VaR) and Expected Shortfall (ES) forecasts are well-specified. We also consider the economic significance of the model’s accuracy through a hedging strategy exercise. Our analysis covers the period from October 10, 2003, to March 28, 2024, including the COVID-19 pandemic. The Realized HAR-GARCH model is found to outperform the Realized GARCH model in volatility forecasting, especially over longer horizons. However, we find this does not directly translate into superior VaR and ES forecasting performance.
dc.description.abstract La inclusión de medidas ya realizadas a modelos GARCH y su impacto favorable sobre su capacidad de pronóstico es un hecho bien documentado en la literatura existente. En este artículo, aplicamos el modelo de Realized HAR-GARCH de Huang et al. (2016) para pronosticar riesgos financieros a la baja. En este ejercicio medimos la precisión del modelo y si las consiguientes medidas de Valor en Riesgo (VeR) y los pronósticos de la caída esperada están bien especificados. Además, medimos la significancia económica de la precisión del modelo con un ejercicio estratégico de cobertura de riesgo. Nuestro análisis cubre el periodo de octubre de 2003 a marzo de 2024, incluyendo la pandemia del COVID-19. Los resultados sugieren que el modelo de Realized HAR-GARCH supera los resultados del Realized GARCH en su habilidad de pronosticar volatilidad, sobre todo cuando el periodo de pronóstico es mayor. Sin embargo, esto no resulta en pronósticos superiores de VeR o caída esperada.
dc.format.mimetype application/pdf
dc.language.iso eng
dc.rights This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License
dc.rights.uri https://creativecommons.org/licenses/by-nc-nd/4.0
dc.subject.other Treball de fi de màster – Curs 2023-2024
dc.title Modelling financial tail risk: multi-step forecasting of value-at-risk and expected shortfall with the realized Har-Garch model
dc.type info:eu-repo/semantics/masterThesis
dc.subject.keyword Volatility forecasting
dc.subject.keyword Value-at-risk
dc.subject.keyword Realized measures
dc.subject.keyword Pronóstico de volatilidad
dc.subject.keyword Valor en riesgo
dc.rights.accessRights info:eu-repo/semantics/openAccess

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse

My Account

Statistics

In collaboration with Compliant to Partaking