Modeling the volatility of returns on commodities: an application and empirical comparison of GARCH and SV models

• Autor: Fernández Prada Saucedo, Jean Pierre
• Formato: tesis de grado
• Fecha de Publicación: 2020
• Institución: Pontificia Universidad Católica del Perú
• Repositorio: PUCP-Tesis
• Lenguaje: inglés
• OAI Identifier: oai:tesis.pucp.edu.pe:20.500.12404/18545
• Enlace del recurso: http://hdl.handle.net/20.500.12404/18545
• Nivel de acceso: acceso abierto
• Materia: Precios--Modelos econométricos; Inversiones--Modelos matemáticos; Productos básicos--Precios--Modelos econométricos; Acciones (Bolsa)--Modelos econométricos; https://purl.org/pe-repo/ocde/ford#5.02.01

Seven GARCH and stochastic volatility (SV) models are compared to model empirically the volatility of returns on four commodities relevant for South America economies: gold, copper, oil, and natural gas. Our results show that SV models outperform GARCH models on average. We find that the best-performing return volatility models are: GARCH-t for gold, SV-t for copper and oil, and SV with leverage effects (SV-L) for natural gas. The inclusion of fat tails and jumps components largely raise the performance of GARCH models, while this contribution is less for SV models. Even, SV models with jumps are usually outperformed by the basic SV model. We also find evidence of a leverage effect in oil and copper, resulting from their dependence on world economic activity; and of an inverse leverage effect in gold and natural gas, consistent with the former's role as safe asset and with uncertainty about the latter's future supply. Additionally, in most cases there is no evidence of an impact of volatility on the mean or MA-type first order autocorrelation.