Markov Chain Monte Carlo Methods for Computing Bayes Factors: A Comparative Review
The DOI matches its Crossref record; 36 of the 36 references it cites were found in a registry or an index.
Abstract
The problem of calculating posterior probabilities for a collection of competing models and associated Bayes factors continues to be a formidable challenge for applied Bayesian statisticians. Current approaches that take advantage of modern Markov chain Monte Carlo computing methods include those that attempt to sample over some form of the joint space created by the model indicators and the parameters for each model, others that sample over the model space alone, and still others that attempt to estimate the marginal likelihood of each model directly (because the collection of these is equivalent to the collection of model probabilities themselves). We review several methods and compare them in the context of three examples: a simple regression example, a more challenging hierarchical longitudinal model, and a binary data latent variable model. We find that the joint model-parameter space search methods perform adequately but can be difficult to program and tune, whereas the marginal likelihood methods often are less troublesome and require less additional coding. Our results suggest that the latter methods may be most appropriate for practitioners working in many standard model choice settings, but the former remain important for comparing models of varying dimension (e.g., multiple changepoint models) or models whose parameters cannot easily be updated in relatively few blocks. We caution, however, that all methods we compared require significant human and computer effort, and this suggests that less formal Bayesian model choice methods may offer a more realistic alternative in many cases.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Estimating the Dimension of a Model
Schwarz, G. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344136 · Cited by 47,033
(2005) · Technometrics · doi:10.1198/tech.2005.s303 · Cited by 17,923
Monte Carlo sampling methods using Markov chains and their applications.
Hastings, W. K. (1970) · Biometrika · doi:10.1093/biomet/57.1.97 · Cited by 15,195 · cited 1×
“In such cases, Bayesian analyses are now typically carried out by using Markov chain Monte Carlo (MCMC) computing methods, such as the Gibbs sampler (Gelfand and Smith 1990) and the Metropolis–Hastings algorithm (Hastings 1970).”
MethodKass, R. E., & Raftery, A. E. (1995) · Journal of the American Statistical Association · doi:10.1080/01621459.1995.10476572 · Cited by 12,558
Markov Chain Monte Carlo in Practice
(1995) · Chapman & Hall/CRC eBooks · doi:10.1201/b14835 · Cited by 7,757
Sampling-based approaches to calculating marginal densities
Gelfand, A. E., & Smith, A. F. M. (1990) · Journal of the American Statistical Association · doi:10.1080/01621459.1990.10476213 · Cited by 6,606 · cited 1×
“In such cases, Bayesian analyses are now typically carried out by using Markov chain Monte Carlo (MCMC) computing methods, such as the Gibbs sampler (Gelfand and Smith 1990) and the Metropolis–Hastings algorithm (Hastings 1970).”
MethodReversible jump Markov chain Monte Carlo computation and Bayesian model determination
GREEN, P. J. (1995) · Biometrika · doi:10.1093/biomet/82.4.711 · Cited by 5,930
Bayesian Analysis of Binary and Polychotomous Response Data
Albert, J. H., & Chib, S. (1993) · Journal of the American Statistical Association · doi:10.1080/01621459.1993.10476321 · Cited by 3,167
Kass, R. E., & Raftery, A. E. (1995) · Journal of the American Statistical Association · doi:10.2307/2291091 · Cited by 2,312
Bayes and Empirical Bayes Methods for Data Analysis
(2001) · Technometrics · doi:10.1198/tech.2001.s608 · Cited by 2,098
Computing Bayes Factors by Combining Simulation and Asymptotic Approximations
Diciccio, T. J., Kass, R. E., Raftery, A., & Wasserman, L. (1997) · Journal of the American Statistical Association · doi:10.1080/01621459.1997.10474045 · Cited by 1,943
Marginal Likelihood from the Gibbs Output
Chib, S. (1995) · Journal of the American Statistical Association · doi:10.1080/01621459.1995.10476635 · Cited by 1,905
On Bayesian Analysis of Mixtures with an Unknown Number of Components (with discussion)
Richardson, S., & Green, P. J. (1997) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/1467-9868.00095 · Cited by 1,898
Bayesian Model Averaging for Linear Regression Models
Raftery, A. E., Hoeting, J. A., & Madigan, D. (1997) · Cited by 1,746
Assessment and Propagation of Model Uncertainty
Draper, D. (1995) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1995.tb02015.x · Cited by 1,599 · cited 1×
“Besides being only single-number summaries of relative model worth, Bayes factors are not interpretable under improper priors on any components of any of the candidate parameters, and they have been criticized on theoretical grounds [see, e.g., Draper (1995) and the associated discussion].”
MethodSampling-Based Approaches to Calculating Marginal Densities
Gelfand, A. E., & Smith, A. F. M. (1990) · Journal of the American Statistical Association · doi:10.2307/2289776 · Cited by 1,504
Bayesian Graphical Models for Discrete Data
Madigan, D., York, J., & Allard, D. (1995) · International Statistical Review / Revue Internationale de Statistique · doi:10.2307/1403615 · Cited by 1,090 · cited 1×
“Unfortunately, such an approach is possible only for fairly special model classes, such as graphical models for discrete data (Madigan and York 1995) and multiple regression models with conjugate priors (Raftery, Madigan, and Hoeting 1997).”
BackgroundMarginal Likelihood From the Metropolis–Hastings Output
Chib, S., & Jeliazkov, I. (2001) · Journal of the American Statistical Association · doi:10.1198/016214501750332848 · Cited by 1,062
Bayesian Model Choice Via Markov Chain Monte Carlo Methods
Carlin, B. P., & Chib, S. (1995) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1995.tb02042.x · Cited by 989
Self-citationMonte Carlo Methods in Bayesian Computation
Chen, M.-H., Shao, Q.-M., & Ibrahim, J. G. (2000) · Springer Series in Statistics · doi:10.1007/978-1-4612-1276-8 · Cited by 911
Analysis of multivariate probit models
Chib, S. (1998) · Biometrika · doi:10.1093/biomet/85.2.347 · Cited by 834
Model choice: a minimum posterior predictive loss approach
Gelfand, A. (1998) · Biometrika · doi:10.1093/biomet/85.1.1 · Cited by 724
Fractional Bayes Factors for Model Comparison
O'Hagan, A. (1995) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1995.tb02017.x · Cited by 671
Model Determination using Predictive Distributions with Implementation via Sampling-Based Methods
Gelfand, A. E., Dey, D. K., & Chang, H. (1992) · Bayesian Statistics 4 · doi:10.1093/oso/9780198522669.003.0009 · Cited by 656
Stephens, M. (2000) · The Annals of Statistics · doi:10.1214/aos/1016120364 · Cited by 524
Hierarchical Spatio-Temporal Mapping of Disease Rates
Waller, L. A., Carlin, B. P., Xia, H., & Gelfand, A. E. (1997) · Journal of the American Statistical Association · doi:10.1080/01621459.1997.10474012 · Cited by 509
Self-citationOn Bayesian model and variable selection using MCMC
Dellaportas, P., Forster, J. J., & Ntzoufras, I. (2002) · Statistics and Computing · doi:10.1023/a:1013164120801 · Cited by 448
Laud, P. W., & Ibrahim, J. G. (1995) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1995.tb02028.x · Cited by 350
Bayesian Model Averaging for Linear Regression Models
Raftery, A. E., Madigan, D., & Hoeting, J. A. (1997) · Journal of the American Statistical Association · doi:10.2307/2291462 · Cited by 294
On the Relationship Between Markov chain Monte Carlo Methods for Model Uncertainty
Godsill, S. J. (2001) · Journal of Computational and Graphical Statistics · doi:10.1198/10618600152627924 · Cited by 281
Bayes Factors: What They are and What They are Not
Lavine, M., & Schervish, M. J. (1999) · The American Statistician · doi:10.1080/00031305.1999.10474443 · Cited by 236
Abrams, D. I., Goldman, A. I., Launer, C., Korvick, J. A., Neaton, J. D., Crane, L. R., Grodesky, M., Wakefield, S., Muth, K., Kornegay, S., et al. (1994) · New England Journal of Medicine · doi:10.1056/nejm199403103301001 · Cited by 214
On MCMC sampling in hierarchical longitudinal models
Chib, S., & Carlin, B. P. (1999) · Statistics and Computing · doi:10.1023/a:1008853808677 · Cited by 185
Self-citationPrediction via Orthogonalized Model Mixing
Clyde, M., Desimone, H., & Parmigiani, G. (1996) · Journal of the American Statistical Association · doi:10.1080/01621459.1996.10476989 · Cited by 145 · cited 2×
“, Draper (1995) and the associated discussion].”
BackgroundBayesian model comparison via jump diffusions
(1995) · Markov Chain Monte Carlo in Practice · doi:10.1201/b14835-18 · Cited by 116
MCMC methods for restoration of nonlinearly distorted autoregressive signals
Troughton, P. T., & Godsill, S. (2001) · Signal Processing · doi:10.1016/s0165-1684(00)00192-4 · Cited by 17
Cited by
Doing Bayesian Data Analysis: A Tutorial with R and BUGS
Barry, J. (2011) · Europe’s Journal of Psychology · doi:10.5964/ejop.v7i4.163 · Cited by 563
Seismic tomography with the reversible jump algorithm
Bodin, T., & Sambridge, M. (2009) · Geophysical Journal International · doi:10.1111/j.1365-246x.2009.04226.x · Cited by 439
Methods and Criteria for Model Selection
Kadane, J. B., & Lazar, N. A. (2004) · Journal of the American Statistical Association · doi:10.1198/016214504000000269 · Cited by 429
von Toussaint, U. (2011) · Reviews of Modern Physics · doi:10.1103/revmodphys.83.943 · Cited by 422
Objective Bayesian Methods for Model Selection: Introduction and Comparison
Berger, J. O., & Pericchi, L. R. (2001) · Institute of Mathematical Statistics Lecture Notes - Monograph Series · doi:10.1214/lnms/1215540968 · Cited by 355
Metadata
- Authors
- Han, C., & Carlin, B. P.
- Venue
- Journal of the American Statistical Association
- Publisher
- Informa UK Limited
- Published
- 2001-09
- Volume
- 96
- Issue
- 455
- Pages
- 1122-1132
- ISSN
- 0162-1459 (ISSN-L)
- 1537-274X
- OpenAlex
- W1974315578
- Cited by
- 275
- References
- 36
- FWCI
- 13.92
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.