Search

Quantification of model risk with Bootstrapping Method

Quantification of model risk with Bootstrapping Method
Recurso electrónico / Electronic resource
Section: Electronic documents
Title: Quantification of model risk with Bootstrapping Method / Li XuAuthor: Xu, Li
Publication: Madrid : Universidad Carlos III de Madrid, 2017Notes: Trabajo Fin de Master del Master en Ciencias Actuariales y Financieras de la Escuela de Postgrado de la Universidad Carlos III de Madrid. Curso 2016-2017Sumario: After the 2007 crisis, it has been widely acknowledged that insufficient understanding of model risk is one of the most critical reasons contributing to the failures of risk model. The concept of model risk intrigues both academic world and financial institutions. Inspired by Boucher et al.(2014), this paper is devoted to finding an applicable method to quantify the model risk. In this paper, model risk is regarded as the excessive frequency of violation and clustering of the violations.
Backtesting method with one-year window is used to search for an optimal capital adjustment for market prevalent VaR models, including GARCH family models, historical simulation and normal distribution method. The properties of the optimal capital adjustment are compared and analyzed and it offers a perspective different from pure statistics measures
Materia / lugar / evento: Cuantificación del riesgo Métodos cuantitativos Cálculo actuarial Evaluación de riesgos Otros autores: Universidad Carlos III de Madrid
Secondary series: Trabajos Fin de Master Other categories: 6