Análisis de los métodos multivariantes para medir el riesgo en una cartera
Loading...
Official URL
Full text at PDC
Publication date
2016
Defense date
2016
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
A la hora de estudiar el valor en riesgo de una cartera, el método univariante puede ser considerado como una sobre simplificación de la realidad. Después de haber experimentado la mayor y más larga crisis financiera de la historia, los mercados buscan una manera efectiva de medir el riesgo. En este estudio haremos un repaso de las principales formas de estimar el VaR y CVaR. El objetivo principal es establecer un indicador cualitativo que nos permita comparar entre los diferentes modelos. Los resultados muestran que la simulación histórica ponderada con un GARCH(1,1) optimiza el control del riesgo.
When estimating the value at risk of a given portfolio, the univariate approch can be an oversimplification of the reality. After having experienced the greatest and the longest financial crisis in documented history the financial market crave for an effective way of measuring risk. In this study we do an overview of the main ways you can estimate and model the VaR and CVaR. The main objective is to do establish a qualitative indicator that could help us to compare between models the models. The findings show that a historical simulation with a GARCH(1,1) approach is the most efficient model.
When estimating the value at risk of a given portfolio, the univariate approch can be an oversimplification of the reality. After having experienced the greatest and the longest financial crisis in documented history the financial market crave for an effective way of measuring risk. In this study we do an overview of the main ways you can estimate and model the VaR and CVaR. The main objective is to do establish a qualitative indicator that could help us to compare between models the models. The findings show that a historical simulation with a GARCH(1,1) approach is the most efficient model.