The likelihood ratio test for the number of components in a mixture with Markov regime

Elisabeth Gassiat; Christine Keribin

ESAIM: Probability and Statistics (2010)

  • Volume: 4, page 25-52
  • ISSN: 1292-8100

Abstract

top
We study the LRT statistic for testing a single population i.i.d. model against a mixture of two populations with Markov regime. We prove that the LRT statistic converges to infinity in probability as the number of observations tends to infinity. This is a consequence of a convergence result of the LRT statistic for a subproblem where the parameters are restricted to a subset of the whole parameter set.

How to cite

top

Gassiat, Elisabeth, and Keribin, Christine. "The likelihood ratio test for the number of components in a mixture with Markov regime." ESAIM: Probability and Statistics 4 (2010): 25-52. <http://eudml.org/doc/197725>.

@article{Gassiat2010,
abstract = { We study the LRT statistic for testing a single population i.i.d. model against a mixture of two populations with Markov regime. We prove that the LRT statistic converges to infinity in probability as the number of observations tends to infinity. This is a consequence of a convergence result of the LRT statistic for a subproblem where the parameters are restricted to a subset of the whole parameter set. },
author = {Gassiat, Elisabeth, Keribin, Christine},
journal = {ESAIM: Probability and Statistics},
keywords = {Likelihood ratio test; hidden Markov model; order of a mixture.; LRT statistic; mixture; Markov regime},
language = {eng},
month = {3},
pages = {25-52},
publisher = {EDP Sciences},
title = {The likelihood ratio test for the number of components in a mixture with Markov regime},
url = {http://eudml.org/doc/197725},
volume = {4},
year = {2010},
}

TY - JOUR
AU - Gassiat, Elisabeth
AU - Keribin, Christine
TI - The likelihood ratio test for the number of components in a mixture with Markov regime
JO - ESAIM: Probability and Statistics
DA - 2010/3//
PB - EDP Sciences
VL - 4
SP - 25
EP - 52
AB - We study the LRT statistic for testing a single population i.i.d. model against a mixture of two populations with Markov regime. We prove that the LRT statistic converges to infinity in probability as the number of observations tends to infinity. This is a consequence of a convergence result of the LRT statistic for a subproblem where the parameters are restricted to a subset of the whole parameter set.
LA - eng
KW - Likelihood ratio test; hidden Markov model; order of a mixture.; LRT statistic; mixture; Markov regime
UR - http://eudml.org/doc/197725
ER -

References

top
  1. L.D. Atwood, A.F. Wilson, J.E. Bailey-Wilson, J.N. Carruth and R.C. Elston, On the distribution of the likelihood ratio test statistic for a mixture of two normal distributions. Comm. Statist. Simulation Comput.25 (1996) 733-740.  Zbl0937.62549
  2. L.E. Baum and T. Petrie, Statistical inference for probabilistic functions of finite state Markov chains. Ann. Math. Stat.37 (1966) 1554-1563.  Zbl0144.40902
  3. P.J. Bickel and Y. Ritov, Inference in hidden Markov models I: Local asymptotic normality in the stationary case. Bernoulli2 (1996) 199-228.  Zbl1066.62535
  4. P.J. Bickel, Y. Ritov and T. Ryden, Asymptotic normality of the maximum-likelihood estimator for general hidden Markov models. Annals of Stat.26 (1998) 1614-1635.  Zbl0932.62097
  5. R.-J. Chuang and N.R. Mendell, The approximate null distribution of the likelihood ratio test for a mixture of two bivariate normal distributions with equal covariance. Comm. Statist. Simulation Comput.26 (1997) 631-648.  Zbl0900.62289
  6. G.A. Churchill, Stochastic models for heterogeneous DNA sequences. Bull. Math. Biology51 (1989) 79-94.  Zbl0662.92012
  7. G. Ciuperca, Sur le test de maximum de vraisemblance pour le mélange de populations. Note aux C.R.A.S., 328, Série I, 4 (1999) 351-358.  
  8. D. Dacunha-Castelle and M. Duflo, Probabilités et statistiques, Tome 2. Masson (1993).  Zbl0535.62003
  9. D. Dacunha-Castelle and E. Gassiat, Estimation of the number of components in a mixture. Bernoulli3 (1997a) 279-299.  Zbl0889.62012
  10. D. Dacunha-Castelle and E. Gassiat, Testing in locally conic models. ESAIM Probab. Statist.1 (1997b).  Zbl1007.62507
  11. D. Dacunha-Castelle and E. Gassiat, Testing the order of a model using locally conic parametrization: Population mixtures and stationary ARMA processes. Ann. Statist.27 (1999) 1178-1209.  Zbl0957.62073
  12. A.P. Dempster, N.M. Laird and D.B. Rubin, Large Maximum-likelihood from incomplete data via the EM algorithm. J. Roy. Statist. Soc. Ser. B39 (1977) 1-38.  Zbl0364.62022
  13. R. Douc and C. Matias, Asymptotics of the Maximum Likelihood Estimator for general Hidden Markov Models (1999) (submitted).  Zbl0987.62018
  14. M. Duflo, Algorithmes stochastiques. Springer (1996).  
  15. Z.D. Feng and C.E. McCulloch, Using bootstrap Likelihood Ratio in Finite Mixture Models. J. Roy. Statist. Soc. Ser. B58 (1996) 609-617.  Zbl0906.62021
  16. L. Finesso, Consistent Estimation of the Order for Markov and Hidden Markov Chains. Ph.D. Thesis, University of Maryland (1990).  
  17. D.R. Fredkin and J.A. Rice, Maximum likelihood estimation and identification directly from single-channel recordings. Proc. Roy. Soc. London Ser. B249 (1992) 125-132.  
  18. P. Hall and C.C. Heyde, Martingale Limit Theory and Its Application. Academic Press (1980).  Zbl0462.60045
  19. J.A. Hartigan, A failure of likelihood ratio asymptotics for normal mixtures, in Proc. Berkeley Conference in Honor of Jerzy Neyman and Jack Kiefer, edited by L.M. Le Cam and R.A. Olshen (1985) 807-810.  
  20. J. Henna, On estimating the number of constituents of a finite mixture of continuous distributions. Ann. Inst. Statist. Math.37 (1985) 235-240.  Zbl0577.62031
  21. J.L. Jensen and N.V. Petersen, Asymptotic normality of the maximum likelihood estimator in state space models. Ann. Statist. 27 (1999) 514-535.  Zbl0952.62023
  22. C. Keribin, Tests de modèles par maximum de vraisemblance, Thèse de l'Université d'Evry-Val d'Essonne (1999).  
  23. C. Keribin, Consistent estimation of the Order of Mixture Models (1997) (submitted).  Zbl1081.62516
  24. B.G. Leroux, Maximum-likelihood estimation for hidden Markov models. Stochastic Process Appl.40 (1992) 127-143.  Zbl0738.62081
  25. B.G. Leroux and M.L. Puterman, Maximum Penalized Likelihood Estimation for Independent and Markov-Dependent Mixture Models. Biometrics48 (1992) 545-558.  
  26. B.G. Lindsay, Mixture models: Theory, Geometry and Applications (1995).  Zbl1163.62326
  27. I.L. Mac Donald and W. Zucchini, Hidden Markov and Other Models for Discrete-valued Time Series. Chapman and Hall (1997).  
  28. G.J. McLachlan, On Bootstrapping the Likelihood Ratio Test Statistic for the Number of Components in a Normal Mixture. Appl. Statist.36 (1987) 318-324.  
  29. L. Mevel, Statistique asymptotique pour les modèles de Markov cachés. Thèse de l'Université de Rennes I (1997).  
  30. L. Mevel and F. LeGland, Exponential forgetting and Geometric Ergodicity in Hidden Markov models. Math. Control Signals Systems (to appear).  Zbl0941.93053
  31. S.P. Meyn and R.L. Tweedie, Markov chains and stochastic stability. Springer-Verlag (1993).  Zbl0925.60001
  32. L.R. Rabiner, A tutorial on hidden Markov models and selected applications in speech recognition. Proc. IEEE77 (1989) 257-284.  
  33. T. Ryden, Estimating the order of hidden Markov models. Statistics26 (1995) 345-354.  Zbl0836.62057
  34. P. Vandekerkhove, Identification de l'ordre des processus ARMA stables. Contribution à l'étude statistique des chaînes de Markov cachées. Thèse de l'Université de Montpellier II (1997).  
  35. A. Van der Vaart, Asymptotic Statistics. Cambridge Ed. (1999).  Zbl0910.62001

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.