A comprehensive review of value at risk methodologies
- Abad Romero, Pilar
- Benito Muela, Sonia
- López Martín, María del Carmen
ISSN: 1988-8767
Year of publication: 2013
Issue: 711
Type: Working paper
More publications in: Notas técnicas: [continuación de Documentos de Trabajo FUNCAS]
Abstract
In this article, we review the full range of methodologies developed to estimate the Value at Risk from standard models to recently proposed methodologies. For these methodologies, we present their relative strengths and weaknesses from theoretical and practical perspectives. From a practical perspective, the empirical literature shows that the approaches based on Extreme Value Theory and Filtered Historical Simulation are the best methods to estimate VaR. Additionally, the Parametric method under the skewed and fat-tail distributions provides results that are promising, especially when the assumption that the standardised returns are independent and identically distributed is abandoned and time variations are considered in the conditional high-order moments.