Probabilistic relabelling strategies for the label switching problem in Bayesian mixture models

@article{Sperrin2010ProbabilisticRS,
  title={Probabilistic relabelling strategies for the label switching problem in Bayesian mixture models},
  author={Matthew Sperrin and Thomas Jaki and Ernst Wit},
  journal={Statistics and Computing},
  year={2010},
  volume={20},
  pages={357-366}
}
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. Most… CONTINUE READING
24 Citations
17 References
Similar Papers

Citations

Publications citing this paper.
Showing 1-10 of 24 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 17 references

Bayesian Inference for Mixture Models via Monte Carlo

  • A. Jasra
  • PhD thesis, Imperial College London,
  • 2005
Highly Influential
6 Excerpts

Estimation of finite mixture distributions through Bayesian sampling

  • J. Diebolt, C. P. Robert
  • J. R. Stat. Soc. Ser. B,
  • 1994
Highly Influential
5 Excerpts

The SEM algorithm: a probabilistic teacher algorithm derived from the em algorithm for the mixture problem

  • G. Celeux, J. Diebolt
  • Comput. Stat. Q.,
  • 1985
Highly Influential
15 Excerpts

Probes of large-scale structure in the Corona Borealis region

  • M. Postman, J. P. Huchra, M. J. Geller
  • Astro. J.,
  • 1986
Highly Influential
3 Excerpts

Similar Papers

Loading similar papers…