Abstract
The label switching problem is caused by the likelihood of a Bayesian mixture model being invariant to permutations of the labels. The permutation can change multiple times between Markov Chain Monte Carlo (MCMC) iterations making it difficult to infer component-specific parameters of the model. Various so-called ‘relabelling’ strategies exist with the goal to ‘undo’ the label switches that have occurred to enable estimation of functions that depend on component-specific parameters. Existing deterministic relabelling algorithms rely upon specifying a loss function, and relabelling by minimising its posterior expected loss. In this paper we develop probabilistic approaches to relabelling that allow for estimation and incorporation of the uncertainty in the relabelling process. Variants of the probabilistic relabelling algorithm are introduced and compared to existing deterministic relabelling algorithms. We demonstrate that the idea of probabilistic relabelling can be expressed in a rigorous framework based on the EM algorithm.
Similar content being viewed by others
References
Aitkin, M.: Likelihood and Bayesian analysis of mixtures. Stat. Model 1, 287–304 (2001)
Celeux, G., Diebolt, J.: The SEM algorithm: a probabilistic teacher algorithm derived from the em algorithm for the mixture problem. Comput. Stat. Q. 2, 73–82 (1985)
Celeux, G., Hurn, M., Robert, C.P.: Computational and inferential difficulties with mixture posterior distributions. J. Am. Stat. Assoc. 95, 957–970 (2000)
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm (with discussion). J. R. Stat. Soc. Ser. B 39, 1–38 (1977)
Diebolt, J., Robert, C.P.: Estimation of finite mixture distributions through Bayesian sampling. J. R. Stat. Soc. Ser. B 56, 363–375 (1994)
Farrar, D.: Approaches to the label-switching problem of classification, based on partition-space relabeling and label-invariant visualization. Technical Report, Statistical Consulting Center and Department of Statistics, Virginia Polytechnic (2006)
Geweke, J.: Interpretation and inference in mixture models: simple MCMC works. Comput. Stat. Data Anal. 51, 3529–3550 (2007)
Hurn, M.A., Justel, A., Robert, C.P.: Estimating mixtures of regressions. J. Comput. Graph. Stat. 12, 55–79 (2003)
Jasra, A.: Bayesian inference for mixture models via Monte Carlo. Ph.D. Thesis, Imperial College London (2005)
Jasra, A., Holmes, C.C., Stephens, D.A.: Markov chain Monte Carlo methods and the label switching problem in Bayesian mixture modelling. Stat. Sci. 20, 50–67 (2005)
Marin, J.M., Mengersen, K.L., Robert, C.P.: Bayesian Modelling and Inference on Mixtures of Distributions. Elsevier, Amsterdam (2005)
McLachlan, G., Peel, D.: Finite Mixture Models. Wiley, New York (2000)
Nobile, A., Fearnside, A.T.: Bayesian finite mixtures with an unknown number of components: the allocation sampler. Stat. Comput. 17(2), 147–162 (2007)
Postman, M., Huchra, J.P., Geller, M.J.: Probes of large-scale structure in the Corona Borealis region. Astron. J. 92, 1238–1246 (1986)
Richardson, S., Green, P.J.: On Bayesian analysis of mixtures with an unknown number of components. J. R. Stat. Soc. Ser. B 59, 758–764 (1997). With discussion
Stephens, M.: Bayesian methods for mixtures of normal distributions. Ph.D. Thesis, University of Oxford (1997a)
Stephens, M.: Discussion of on Bayesian analysis of mixtures with an unknown number of components. J. R. Stat. Soc. Ser. B 59, 768–769 (1997b)
Stephens, M.: Dealing with label-switching in mixture models. J. R. Stat. Soc. Ser. B 62, 795–809 (2000)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sperrin, M., Jaki, T. & Wit, E. Probabilistic relabelling strategies for the label switching problem in Bayesian mixture models. Stat Comput 20, 357–366 (2010). https://doi.org/10.1007/s11222-009-9129-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-009-9129-8