Exact Bayesian inference for the Bingham distribution

This paper is concerned with making Bayesian inference from data that are assumed to be drawn from a Bingham distribution. A barrier to the Bayesian approach is the parameter-dependent normalising constant of the Bingham distribution, which, even when it can be evaluated or accurately approximated,...

Full description

Bibliographic Details
Main Authors: Fallaize, Christopher J., Kypraios, Theodore
Format: Article
Published: Springer 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/35016/
_version_ 1848794984045608960
author Fallaize, Christopher J.
Kypraios, Theodore
author_facet Fallaize, Christopher J.
Kypraios, Theodore
author_sort Fallaize, Christopher J.
building Nottingham Research Data Repository
collection Online Access
description This paper is concerned with making Bayesian inference from data that are assumed to be drawn from a Bingham distribution. A barrier to the Bayesian approach is the parameter-dependent normalising constant of the Bingham distribution, which, even when it can be evaluated or accurately approximated, would have to be calculated at each iteration of an MCMC scheme, thereby greatly increasing the computational burden. We propose a method which enables exact (in Monte Carlo sense) Bayesian inference for the unknown parameters of the Bingham distribution by completely avoiding the need to evaluate this constant. We apply the method to simulated and real data, and illustrate that it is simpler to implement, faster, and performs better than an alternative algorithm that has recently been proposed in the literature
first_indexed 2025-11-14T19:24:52Z
format Article
id nottingham-35016
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:24:52Z
publishDate 2016
publisher Springer
recordtype eprints
repository_type Digital Repository
spelling nottingham-350162020-05-04T20:04:32Z https://eprints.nottingham.ac.uk/35016/ Exact Bayesian inference for the Bingham distribution Fallaize, Christopher J. Kypraios, Theodore This paper is concerned with making Bayesian inference from data that are assumed to be drawn from a Bingham distribution. A barrier to the Bayesian approach is the parameter-dependent normalising constant of the Bingham distribution, which, even when it can be evaluated or accurately approximated, would have to be calculated at each iteration of an MCMC scheme, thereby greatly increasing the computational burden. We propose a method which enables exact (in Monte Carlo sense) Bayesian inference for the unknown parameters of the Bingham distribution by completely avoiding the need to evaluate this constant. We apply the method to simulated and real data, and illustrate that it is simpler to implement, faster, and performs better than an alternative algorithm that has recently been proposed in the literature Springer 2016-01 Article PeerReviewed Fallaize, Christopher J. and Kypraios, Theodore (2016) Exact Bayesian inference for the Bingham distribution. Statistics and Computing, 26 (1). pp. 349-360. ISSN 1573-1375 Directional statistics; Bayesian inference; Markov Chain Monte Carlo; Doubly intractable distributions http://link.springer.com/article/10.1007/s11222-014-9508-7 doi:10.1007/s11222-014-9508-7 doi:10.1007/s11222-014-9508-7
spellingShingle Directional statistics; Bayesian inference; Markov Chain Monte Carlo; Doubly intractable distributions
Fallaize, Christopher J.
Kypraios, Theodore
Exact Bayesian inference for the Bingham distribution
title Exact Bayesian inference for the Bingham distribution
title_full Exact Bayesian inference for the Bingham distribution
title_fullStr Exact Bayesian inference for the Bingham distribution
title_full_unstemmed Exact Bayesian inference for the Bingham distribution
title_short Exact Bayesian inference for the Bingham distribution
title_sort exact bayesian inference for the bingham distribution
topic Directional statistics; Bayesian inference; Markov Chain Monte Carlo; Doubly intractable distributions
url https://eprints.nottingham.ac.uk/35016/
https://eprints.nottingham.ac.uk/35016/
https://eprints.nottingham.ac.uk/35016/