Skip to main content
SpringerLink
Log in

An expectation maximization algorithm for the hidden markov models with multiparameter student-t observations

  • Original Paper
  • Published:
Computational Statistics Aims and scope Submit manuscript

Abstract

Hidden Markov models are a class of probabilistic graphical models used to describe the evolution of a sequence of unknown variables from a set of observed variables. They are statistical models introduced by Baum and Petrie in Baum (JMA 101:789–810) and belong to the class of latent variable models. Initially developed and applied in the context of speech recognition, they have attracted much attention in many fields of application. The central objective of this research work is upon an extension of these models. More accurately, we define multiparameter hidden Markov models, using multiple observation processes and the Riesz distribution on the space of symmetric matrices as a natural extension of the gamma one. Some basic related properties are discussed and marginal and posterior distributions are derived. We conduct the Forward-Backward dynamic programming algorithm and the classical Expectation Maximization algorithm to estimate the global set of parameters. Using simulated data, the performance of these estimators is conveniently achieved by the Matlab program. This allows us to assess the quality of the proposed estimators by means of the mean square errors between the true and the estimated values.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

References

  • Andersson SA, Klein T (2010) On Riesz and Wishart distributions associated with decomposable undirected graphs. J Multivar Anal 101(4):789–810

    Article  MathSciNet  MATH  Google Scholar 

  • Baum LE, Petrie T (1966) Statistical inference for probabilistic functions of finite state Markov chains. Ann math Stat 37(6):1554–1563

    Article  MathSciNet  MATH  Google Scholar 

  • Brunel N, Lapuyade-Lahorgue J, Pieczynski W (2010) Modeling and unsupervised classification of multivariate hidden Markov chains with copulas. IEEE Trans on Automatic Control 55(2):338–349

    Article  MathSciNet  MATH  Google Scholar 

  • Cappé O, Moulines E, Rydén T (2005) Inference in hidden Markov models. Springer-VerlagBerlin, Heidelberg

    Book  MATH  Google Scholar 

  • Cho W, Na MH, Kim S (2016a) Human action recognition using variational Bayesian HMM with Dirichlet process mixture of Gaussian Wishart emission model. World Acad Sci Eng Technol Int J Comput Elect Autom Control Information Eng 10:2025–2031

    Google Scholar 

  • Cho W, Kim S, Park S (2016b) Human action recognition using variational Bayesian hidden Markov model with Gaussian-Wishart emission mixture model. In: international conference on machine learning and cybernetics (ICMLC) 201–206

  • Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Statist 39(1):1–38

    MathSciNet  MATH  Google Scholar 

  • Devijver PA, Dekesel M (1988) Champs aléatoires de Pickard et modélisation d’images digitales. Trait Signal 5(5):131–150

    Google Scholar 

  • Díaz-García JA (2014) On Riesz distribution. Metrika 77:469–481

    Article  MathSciNet  MATH  Google Scholar 

  • Douc R, Fort G, Moulines E, Priouret P (2009) Forgetting of the initial distribution for hidden Markov models. Stoch Process Their Appl 119(4):1235–1256

    Article  MathSciNet  MATH  Google Scholar 

  • Dunn JC (1973) A Fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters. J Cybern 3:32–57

    Article  MathSciNet  MATH  Google Scholar 

  • El-Emary IMM, Fezari M, Atoui H (2011) Hidden Markov model/Gaussian mixture models (HMM/GMM) based voice command system: A way to improve the control of remotely operated robot arm TR45. Scientific Research and Essays 6(2):341–350

    Google Scholar 

  • Fan W, Yang L, Bouguila N, Chen Y (2020) Sequentially spherical data modeling with hidden Markov models and its application to fMRI data analysis. Knowl Based Syst 206:106341

    Article  Google Scholar 

  • Faraut J, Korányi F (1994) Analysis on symmetric cones. Oxford University Press

    Book  MATH  Google Scholar 

  • Ghahramani Z (2001) An Introduction to hidden Markov models and Bayesian networks. Int J Pattern Recognit Artif Intell 15:9–42

    Article  Google Scholar 

  • Ghorbel E, Louati M (2019) The multiparameter t’distribution. Filomat 33(13):4137–4150

    Article  MathSciNet  MATH  Google Scholar 

  • Ghorbel E, Kammoun K, Louati M (2020) Bayesian estimation of the precision matrix with monotone missing data. Lith Math J 60(4):470–481

    Article  MathSciNet  MATH  Google Scholar 

  • Ghorbel E, Kammoun K, Louati M, Sallem A (2022) Estimation of the parameters of a Wishart extension on symmetric matrices. J Korean Statist Soc 51:1071–1089

    Article  MathSciNet  MATH  Google Scholar 

  • Gindikin SG (1964) Analysis in homogeneous domains. Russian Math Surveys 19(4):1–89

    Article  MathSciNet  MATH  Google Scholar 

  • Haff LR (1982) Identities for the inverse Wishart distribution with computational results in linear and quadratic discrimination. Sankhyā Ser B 44(3):245–258

    MathSciNet  MATH  Google Scholar 

  • Hassairi A, Louati M (2009) Multivariate stable exponential families and Tweedie scale. J Stat Plan Inference 139:143–158

    Article  MathSciNet  MATH  Google Scholar 

  • Hidot S, Saint-Jean C (2010) An Expectation-Maximization algorithm for the Wishart mixture model: Application to movement clustering. Patt Recognit Lett 31:2318–2324

    Article  Google Scholar 

  • Higgins C, Vidaurre D, Kolling N, Liu Y, Behrens T, Woolrich M (2022) Spatiotemporally resolved multivariate pattern analysis for M/EEG. Hum Brain Mapp 43(10): 3062–3085

    Article  Google Scholar 

  • Ishi H (2000) Positive Riesz distributions on homogeneous cones. J Math Soc Japan 52(1), 161–186

    Article  MathSciNet  MATH  Google Scholar 

  • Kaleh GK, Vallet R (1994) Joint parameter estimation and symbol detection for linear or nonlinear uknown channels. IEEE Trans Commun 42(7), 406–413

    MATH  Google Scholar 

  • Kammoun K, Louati M, Masmoudi A (2017) Maximum likelihood estimator of the scale parameter for the Riesz distribution. Stat Probabil Lett 126:127–131

    Article  MathSciNet  MATH  Google Scholar 

  • Li J, Lee JY, Liao L (2021) A new algorithm to train hidden Markov models for biological sequences with partial labels. BMC Bioinform 22:162

    Article  Google Scholar 

  • Lin P (1972) Some characterizations of the multivariate t distribution. J Multivar Anal 2:339–344

    Article  MathSciNet  Google Scholar 

  • Louati M (2013) Mixture of the Riesz distribution with respect to the generalized multivariate gamma distribution. J Korean Statist Soc 42:83–93

    Article  MathSciNet  MATH  Google Scholar 

  • Louati M, Masmoudi A (2015) Moment for the inverse Riesz distributions. Stat Probabil lett 102:30–37

    Article  MathSciNet  MATH  Google Scholar 

  • Mesa A, Basterrech A, Guerberoff G, Alveraz-Valin F (2015) Hidden Markov models for gene sequence classification. Pattern Anal Appl 19:793–805

    Article  Google Scholar 

  • Osatohanmwen P, Omotayo-Tomo MS, Oyegue FO, Mazona V, Ewere F, Bilesanmi A, Osawe NL, (2023) A note on hidden Markov models with application to criminal intelligence. J Appl Sci Environ Manag 27(2):277–282

    Google Scholar 

  • Petropoulos A, Chatzis SP, Xanthopoulos S (2016) A novel corporate credit rating system based on Student’s-t hidden Markov models. Expert Syst Appl 53:87–105

    Article  Google Scholar 

  • Qian W, Titterington DM (1989) On the use of Gibbs Markov chain models in the analysis of images based on second-order pairwise interactive distributions. J Appl Stat 6(2), 267–282

    Article  Google Scholar 

  • Rösler M (2020) Riesz distributions and Laplace transform in the Dunkl setting of type A. J Funct Anal 278(12):108506

    Article  MathSciNet  MATH  Google Scholar 

  • Stigler J, Ziegler F, Gieseke A, Gebhardt JCM, Rief M (2011) The complex folding network of single calmodulin molecule. Science 334(6055): 512–516

    Article  Google Scholar 

  • Suleiman D, Awajan A, Al Etaiwi W (2017) The use of hidden Markov model in natural ARABIC language processing: a survey. Procedia Comput Sci 113:240–247

    Article  Google Scholar 

  • Tian F, Zhou Q, Yang C (2020) Gaussian mixture model-hidden Markov model based nonlinear equalizer for optical fiber transmission. Opt Express 28:9728–9737

    Article  Google Scholar 

  • Veleva E (2009) Testing a Normal covariance matrix for small samples with monotone missing data. Appl math sci 3(54):2671–2679

    MathSciNet  MATH  Google Scholar 

  • Viterbi A (1967) Error bounds for convolutional codes and an asymptotically optimum decoding algorithm. IEEE Trans Inform Theory 13:260–269

    Article  MATH  Google Scholar 

  • von Rosen D (1988) Moments for the inverted Wishart distribution. Scand J Stat 15:97–109

    MathSciNet  MATH  Google Scholar 

  • Xuan G, Wei Zhang W, Chai P (2001) EM algorithms of Gaussian mixture model and hidden Markov model. Int Conf Image Process 1:145–148

    Google Scholar 

  • Zhang H, Zhang W, Palazoglu A, Sun W (2012) Prediction of ozone levels using a hidden Markov model (HMM) with gamma distribution. Atmosph Environ 62:64–73

    Article  Google Scholar 

Download references

Author information

Author notes

  1. Emna Ghorbel and Mahdi Louati have contributed equally to this work.

Authors and Affiliations

  1. ESPRIT School of Engineering, Z.I. Chotrana II, 2083, Ariana, Tunis, Tunisia

    Emna Ghorbel

  2. Laboratory of Signals, Systems, Artificial Intelligence and Networks (SM@RTS), Digital Research Center of Sfax, 3021, Sakiet Ezzit, Sfax, Tunisia

    Mahdi Louati

Authors
  1. Emna Ghorbel
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Mahdi Louati
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mahdi Louati.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file 1 (pdf 241 KB)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ghorbel, E., Louati, M. An expectation maximization algorithm for the hidden markov models with multiparameter student-t observations. Comput Stat (2023). https://doi.org/10.1007/s00180-023-01432-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00180-023-01432-7

Keywords

Search

Navigation