Evaluation of volatility models with long memory: Evidence from Peru

• Autor: Briones Zúñiga, José Luis
• Formato: artículo
• Fecha de Publicación: 2020
• Institución: Universidad Nacional Mayor de San Marcos
• Repositorio: Revistas - Universidad Nacional Mayor de San Marcos
• Lenguaje: español
• OAI Identifier: oai:ojs.csi.unmsm:article/19342
• Enlace del recurso: https://revistasinvestigacion.unmsm.edu.pe/index.php/matema/article/view/19342
• Nivel de acceso: acceso abierto
• Materia: volatily; GARCH; FIGARCH.; volatilidad; FIGARCH

The objective of the study is to compare long memory models to model exchange rate volatility. For this objective, the nominal sol / dollar exchange rate is used, covering the periods from July 19, 1999 to November 19, 2013. Essentially, it seeks to examine the prediction capacity between long memory models and hyperbolic behavior of the autocorrelations given by FIGARCH, HYGARCH and IGARCH and concluding that the FIGARCH model (1,0,637,1) using a t-student distribution has a better predictive capacity. The prediction of exchange rate volatility in the case of Peru is structurally important in the calculation of Value at Risk (VaR) and in risk management.