Adaptive Polar Sampling
with an application to a Bayes measure of Value-at-Risk

Charles S. BOS1, Luc Bauwens and Herman K. van Dijk
Econometric and Tinbergen Institutes, Erasmus University Rotterdam,
CORE, Université catholique de Louvain and
Econometric Institute, Erasmus University Rotterdam

October 21, 1999

Abstract

Adaptive Polar Sampling (APS) is proposed as a Markov chain Monte Carlo method for Bayesian analysis of models with ill-behaved posterior distributions. In order to sample efficiently from such a distribution, a location-scale transformation and a transformation to polar coordinates are used. After the transformation to polar coordinates, a Metropolis-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 near non-identifiability, strong correlation, and bimodality, APS compares favourably with the standard Metropolis-Hastings sampler in terms of parsimony and robustness. APS is applied within a Bayesian analysis of a GARCH-mixture model which is used for the evaluation of the Value-at-Risk of the return of the Dow Jones stock index.

JEL classification: C11, C15, C63
Keywords: Markov chain Monte Carlo, simulation, polar coordinates, GARCH, ill-behaved posterior, Value-at-Risk


Footnotes:

1 Correspondence to Charles S. Bos, Tinbergen Institute, Erasmus University Rotterdam, Burg. Oudlaan 50, NL-3062 PA  Rotterdam, The Netherlands. Email: cbos@feweb.vu.nl, URL: http://www.tinbergen.nl/~cbos/. Support from HCM grant ERBCHRXCT 940514 is gratefully acknowledged. We thank Michel Lubrano and participants at ESEM'99 for helpful comments on earlier versions of this paper.


File translated from TEX by TTH, version 2.50.


[Main page] [Top] [Next: Hedging Currency Risk]
Last change: 7/12/2001