Home > Research > Publications & Outputs > Note on Posterior Inference for the Bingham Dis...

Electronic data

  • Note on Posterior Inference for the Bingham Distribution

    Rights statement: This is an Accepted Manuscript of an article published by Taylor & Francis in Communication in Statistics - Theory and Methods on 30/06/2017, available online: http://www.tandfonline.com/10.1080/03610926.2017.1346805

    Accepted author manuscript, 591 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

  • Note on the Bingham distribution

    Rights statement: This is an Accepted Manuscript of an article published by Taylor & Francis in Communication in Statistics - Theory and Methods on 30/06/2017, available online: http://www.tandfonline.com/10.1080/03610926.2017.1346805

    Accepted author manuscript, 248 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Note on Posterior Inference for the Bingham Distribution

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published
<mark>Journal publication date</mark>2018
<mark>Journal</mark>Communications in Statistics - Theory and Methods
Issue number12
Volume47
Number of pages7
Pages (from-to)3022-3028
Publication StatusPublished
Early online date30/06/17
<mark>Original language</mark>English

Abstract

The properties of high-dimensional Bingham distributions have been studied by Kume and Walker. Fallaize and Kypraios propose Bayesian inference for the Bingham distribution and they use developments in Bayesian computation for distributions with doubly intractable normalising constants (Møller et al. 2006 ; Murray et al. 2006. However, they rely heavily on two Metropolis updates that they need to tune. In this paper we propose instead model selection with the marginal likelihood.