Bernoulli

  • Bernoulli
  • Volume 20, Number 4 (2014), 2039-2075.

About the posterior distribution in hidden Markov models with unknown number of states

Elisabeth Gassiat and Judith Rousseau

Full-text: Open access

Abstract

We consider finite state space stationary hidden Markov models (HMMs) in the situation where the number of hidden states is unknown. We provide a frequentist asymptotic evaluation of Bayesian analysis methods. Our main result gives posterior concentration rates for the marginal densities, that is for the density of a fixed number of consecutive observations. Using conditions on the prior, we are then able to define a consistent Bayesian estimator of the number of hidden states. It is known that the likelihood ratio test statistic for overfitted HMMs has a nonstandard behaviour and is unbounded. Our conditions on the prior may be seen as a way to penalize parameters to avoid this phenomenon. Inference of parameters is a much more difficult task than inference of marginal densities, we still provide a precise description of the situation when the observations are i.i.d. and we allow for $2$ possible hidden states.

Article information

Source
Bernoulli, Volume 20, Number 4 (2014), 2039-2075.

Dates
First available in Project Euclid: 19 September 2014

Permanent link to this document
https://projecteuclid.org/euclid.bj/1411134453

Digital Object Identifier
doi:10.3150/13-BEJ550

Mathematical Reviews number (MathSciNet)
MR3263098

Zentralblatt MATH identifier
1302.62183

Keywords
Bayesian statistics hidden Markov models number of components order selection posterior distribution

Citation

Gassiat, Elisabeth; Rousseau, Judith. About the posterior distribution in hidden Markov models with unknown number of states. Bernoulli 20 (2014), no. 4, 2039--2075. doi:10.3150/13-BEJ550. https://projecteuclid.org/euclid.bj/1411134453


Export citation

References

  • [1] Boys, R.J. and Henderson, D.A. (2004). A Bayesian approach to DNA sequence segmentation. Biometrics 60 573–588. With discussions and a reply by the author.
  • [2] Cappé, O., Moulines, E. and Rydén, T. (2004). Hidden Markov Models. New York: Springer.
  • [3] Chambaz, A., Garivier, A. and Gassiat, E. (2009). A minimum description length approach to hidden Markov models with Poisson and Gaussian emissions. Application to order identification. J. Statist. Plann. Inference 139 962–977.
  • [4] de Gunst, M.C.M. and Shcherbakova, O. (2008). Asymptotic behavior of Bayes estimators for hidden Markov models with application to ion channels. Math. Methods Statist. 17 342–356.
  • [5] Douc, R., Moulines, É. and Rydén, T. (2004). Asymptotic properties of the maximum likelihood estimator in autoregressive models with Markov regime. Ann. Statist. 32 2254–2304.
  • [6] Gassiat, E. (2002). Likelihood ratio inequalities with applications to various mixtures. Ann. Inst. Henri Poincaré Probab. Stat. 38 897–906.
  • [7] Gassiat, E. and Boucheron, S. (2003). Optimal error exponents in hidden Markov models order estimation. IEEE Trans. Inform. Theory 49 964–980.
  • [8] Gassiat, E. and van Handel, R. (2014). The local geometry of finite mixtures. Trans. Amer. Math. Soc. 366 1047–1072.
  • [9] Gassiat, E. and Keribin, C. (2000). The likelihood ratio test for the number of components in a mixture with Markov regime. ESAIM Probab. Statist. 4 25–52.
  • [10] Ghosal, S. and van der Vaart, A. (2007). Convergence rates of posterior distributions for non-i.i.d. observations. Ann. Statist. 35 192–223.
  • [11] Ghosh, J.K. and Ramamoorthi, R.V. (2003). Bayesian Nonparametrics. Springer Series in Statistics. New York: Springer.
  • [12] Green, P.J. and Richardson, S. (2002). Hidden Markov models and disease mapping. J. Amer. Statist. Assoc. 97 1055–1070.
  • [13] Leroux, B. and Putterman, M. (1992). Maximum-penalised-likelihood estimation for independent and Markov dependent mixture models. Biometrics 48 545–558.
  • [14] MacDonald, I.L. and Zucchini, W. (1997). Hidden Markov and Other Models for Discrete-Valued Time Series. Monographs on Statistics and Applied Probability 70. London: Chapman & Hall.
  • [15] McGrory, C.A. and Titterington, D.M. (2009). Variational Bayesian analysis for hidden Markov models. Aust. N. Z. J. Stat. 51 227–244.
  • [16] Nur, D., Allingham, D., Rousseau, J., Mengersen, K.L. and McVinish, R. (2009). Bayesian hidden Markov model for DNA sequence segmentation: A prior sensitivity analysis. Comput. Statist. Data Anal. 53 1873–1882.
  • [17] Richardson, S. and Green, P.J. (1997). On Bayesian analysis of mixtures with an unknown number of components. J. Roy. Statist. Soc. Ser. B 59 731–792.
  • [18] Rio, E. (2000). Inégalités de Hoeffding pour les fonctions lipschitziennes de suites dépendantes. C. R. Acad. Sci. Paris Sér. I Math. 330 905–908.
  • [19] Robert, C.P., Rydén, T. and Titterington, D.M. (2000). Bayesian inference in hidden Markov models through the reversible jump Markov chain Monte Carlo method. J. R. Stat. Soc. Ser. B Stat. Methodol. 62 57–75.
  • [20] Rousseau, J. and Mengersen, K. (2011). Asymptotic behaviour of the posterior distribution in overfitted mixture models. J. R. Stat. Soc. Ser. B Stat. Methodol. 73 689–710.
  • [21] Rydén, T., Terasvirta, T. and Asbrink, S. (1998). Stylized facts of daily return series and the hidden Markov model. J. Appl. Econometrics 13 217–244.
  • [22] Spezia, L. (2010). Bayesian analysis of multivariate Gaussian hidden Markov models with an unknown number of regimes. J. Time Series Anal. 31 1–11.
  • [23] Zucchini, W. and MacDonald, I.L. (2009). Hidden Markov Models for Time Series: An Introduction Using R. Monographs on Statistics and Applied Probability 110. Boca Raton, FL: CRC Press.