Publication: Sample size, skewness and leverage effects in value at risk and expected shortfall estimation
Loading...
Files
Official URL
Full text at PDC
Publication Date
2018-01-30
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Complutense de Madrid
Citations
Exportar
Abstract
La estimación de las medidas de riesgo es un area de gran importancia en la industria financiera. Las medidas de riesgo juegan un papel principal en la gesti on del riesgo y en el c alculo del capital requerido. El documento Basilea III [13] ha sugerido medir el riesgo en condiciones de tensi on mediante la P erdida Esperada (ES), en lugar del Valor en Riesgo (VaR), a un nivel de con anza del 97:5%. Este cambio viene motivado por las atractivas propiedades te oricas del ES como medida de riesgo y por las limitaciones del VaR. En particular, el VaR no captura el "riesgo de cola". En esta transici on, el principal reto al que se enfrentan las instituciones nancieras es la falta de disponibilidad de herramientas sencillas para la evaluaci on de las predicciones del ES, esto es, backtesting del ES. El objetivo de la tesis es comparar la performance de una variedad de modelos para la estimaci on del VaR y del ES para un conjunto de activos de diferente naturaleza: ndices de mercado, acciones, bonos, tipos de cambio y materias primas. A lo largo de la tesis, entendemos por \modelo" VaR o por \modelo" ES una especi caci on dada por un modelo de volatilidad condicional combinado con una distribuci on de probabilidad que suponemos siguen las innovaciones estandarizadas...
The estimation of risk measures is an area of highest importance in the financial industry. Risk measures play a major role in the risk-management and in the computation of regulatory capital. The Basel III document [13] has suggested to shift from Value-at-Risk (VaR) into Expected Shortfall (ES) as a risk measure and to consider stressed scenarios at a new con dence level of 97:5%. This change is motivated by the appealing theoretical properties of ES as a measure of risk and the poor properties of VaR. In particular, VaR fails to control for \tail risk". In this transition, the major challenge faced by nancial institutions is the unavailability of simple tools for evaluation of ES forecasts (i.e. backtesting ES) The objective of this thesis is to compare the performance of a variety of models for VaR and ES estimation for a collection of assets of di erent nature: stock indexes, individual stocks, bonds, exchange rates, and commodities. Throughout the thesis, by a VaR or an ES \model" is meant a given speci cation for conditional volatility, combined with an assumption on the probability distribution of return innovations...
The estimation of risk measures is an area of highest importance in the financial industry. Risk measures play a major role in the risk-management and in the computation of regulatory capital. The Basel III document [13] has suggested to shift from Value-at-Risk (VaR) into Expected Shortfall (ES) as a risk measure and to consider stressed scenarios at a new con dence level of 97:5%. This change is motivated by the appealing theoretical properties of ES as a measure of risk and the poor properties of VaR. In particular, VaR fails to control for \tail risk". In this transition, the major challenge faced by nancial institutions is the unavailability of simple tools for evaluation of ES forecasts (i.e. backtesting ES) The objective of this thesis is to compare the performance of a variety of models for VaR and ES estimation for a collection of assets of di erent nature: stock indexes, individual stocks, bonds, exchange rates, and commodities. Throughout the thesis, by a VaR or an ES \model" is meant a given speci cation for conditional volatility, combined with an assumption on the probability distribution of return innovations...
Description
Tesis inédita de la Universidad Complutense de Madrid, Facultad de Ciencias Económicas y Empresariales, Departamento de Fundamentos del Análisis Económico II (Economía Cuantitativa), leída el 20-09-2017