A general framework for the parametrization of hierarchical models
(2007) In Statistical Science 22(1). p.59-73- Abstract
- In this paper, we describe centering and noncentering methodology as complementary techniques for use in parametrization of broad classes of hierarchical models, with a view to the construction of effective MCMC algorithms for exploring posterior distributions from these models. We give a clear qualitative understanding as to when centering and noncentering work well, and introduce theory concerning the convergence time complexity of Gibbs samplers using centered and noncentered parametrizations. We give general recipes for the construction of noncentered parametrizations, including an auxiliary variable technique called the state-space expansion technique. We also describe partially noncentered methods, and demonstrate their use in... (More)
- In this paper, we describe centering and noncentering methodology as complementary techniques for use in parametrization of broad classes of hierarchical models, with a view to the construction of effective MCMC algorithms for exploring posterior distributions from these models. We give a clear qualitative understanding as to when centering and noncentering work well, and introduce theory concerning the convergence time complexity of Gibbs samplers using centered and noncentered parametrizations. We give general recipes for the construction of noncentered parametrizations, including an auxiliary variable technique called the state-space expansion technique. We also describe partially noncentered methods, and demonstrate their use in constructing robust Gibbs sampler algorithms whose convergence properties are not overly sensitive to the data. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/688610
- author
- Papaspiliopoulos, Omiros ; Roberts, Gareth O. and Sköld, Martin LU
- organization
- publishing date
- 2007
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- latent stochastic processes, parametrization, hierarchical models, MCMC
- in
- Statistical Science
- volume
- 22
- issue
- 1
- pages
- 59 - 73
- publisher
- IMS
- external identifiers
-
- wos:000249036700007
- scopus:34249101736
- ISSN
- 0883-4237
- DOI
- 10.1214/088342307000000014
- language
- English
- LU publication?
- yes
- id
- c9c02cec-8ecc-4d00-8646-4ad43806762f (old id 688610)
- date added to LUP
- 2016-04-01 16:41:16
- date last changed
- 2022-03-15 02:06:29
@article{c9c02cec-8ecc-4d00-8646-4ad43806762f, abstract = {{In this paper, we describe centering and noncentering methodology as complementary techniques for use in parametrization of broad classes of hierarchical models, with a view to the construction of effective MCMC algorithms for exploring posterior distributions from these models. We give a clear qualitative understanding as to when centering and noncentering work well, and introduce theory concerning the convergence time complexity of Gibbs samplers using centered and noncentered parametrizations. We give general recipes for the construction of noncentered parametrizations, including an auxiliary variable technique called the state-space expansion technique. We also describe partially noncentered methods, and demonstrate their use in constructing robust Gibbs sampler algorithms whose convergence properties are not overly sensitive to the data.}}, author = {{Papaspiliopoulos, Omiros and Roberts, Gareth O. and Sköld, Martin}}, issn = {{0883-4237}}, keywords = {{latent stochastic processes; parametrization; hierarchical models; MCMC}}, language = {{eng}}, number = {{1}}, pages = {{59--73}}, publisher = {{IMS}}, series = {{Statistical Science}}, title = {{A general framework for the parametrization of hierarchical models}}, url = {{http://dx.doi.org/10.1214/088342307000000014}}, doi = {{10.1214/088342307000000014}}, volume = {{22}}, year = {{2007}}, }