Cover Image

Adaptive Polar Sampling with an Application to a Bayes Measure of Value-at-Risk

Adaptive Polar Sampling (APS) is proposed as a Markov chain Monte Carlomethod for Bayesian analysis of models with ill-behaved posteriordistributions. In order to sample efficiently from such a distribution,a location-scale transformation and a transformation to polarcoordinates are used. After the transformation to polar coordinates, aMetropolis-Hastings algorithm is applied to sample directions and,conditionally on these, distances are generated by inverting the CDF.A sequential procedure is applied to update the location and scale.Tested on a set of canonical models that feature nearnon-identifiability, strong correlation, and bimodality, APS comparesfavourably with the standard Metropolis-Hastings sampler in terms ofparsimony and robustness. APS is applied within a Bayesian analysisof a GARCH-mixture model which is used for the evaluation of theValue-at-Risk of the return of the Dow Jones stock index.
Year of publication:
1999
Authors: Bauwens, Luc ; Bos, Charles S. ; van Dijk, Herman K.
Publisher: Amsterdam and Rotterdam : Tinbergen Institute
Subject: Stichprobenverfahren | Theorie | Markov chain Monte Carlo | simulation | polar coordinates | GARCH | ill-behaved posterior | Value-at-Risk
Series:
Type of publication: Book / Working Paper
Type of publication (narrower categories): Working Paper
Language: English
Other identifiers:
832629189 [GVK]
hdl:10419/85730 [Handle]
RePEc:dgr:uvatin:19990082 [RePEc]
Classification: C11 - Bayesian Analysis ; C15 - Statistical Simulation Methods; Monte Carlo Methods ; C63 - Computational Techniques
Source:
Persistent link: https://ebvufind01.dmz1.zbw.eu/10010324702
Similar items by subject
Similar items by person