Annals of Statistics

Convergence of a stochastic approximation version of the EM algorithm

Bernard Delyon, Marc Lavielle, and Eric Moulines

Full-text: Open access

PDF File (239 KB)

Abstract

The expectation-maximization (EM) algorithm is a powerful computational technique for locating maxima of functions. It is widely used in statistics for maximum likelihood or maximum a posteriori estimation in incomplete data models. In certain situations, however, this method is not applicable because the expectation step cannot be performed in closed form. To deal with these problems, a novel method is introduced, the stochastic approximation EM (SAEM), which replaces the expectation step of the EM algorithm by one iteration of a stochastic approximation procedure. The convergence of the SAEM algorithm is established under conditions that are applicable to many practical situations. Moreover, it is proved that, under mild additional conditions, the attractive stationary points of the SAEM algorithm correspond to the local maxima of the function presented to support our findings.

Article information

Source
Ann. Statist., Volume 27, Number 1 (1999), 94-128.

Dates
First available in Project Euclid: 5 April 2002

Permanent link to this document
https://projecteuclid.org/euclid.aos/1018031103

Digital Object Identifier
doi:10.1214/aos/1018031103

Mathematical Reviews number (MathSciNet)
MR1701103

Zentralblatt MATH identifier
0932.62094

Subjects
Primary: 65U05 62F10: Point estimation
Secondary: 62M30: Spatial processes 60K35: Interacting random processes; statistical mechanics type models; percolation theory [See also 82B43, 82C43]

Keywords
Incomplete data optimization maximum likelihood missing data Monte Carlo algorithm EM algorithm simulation stochastic algorithm

Citation

Delyon, Bernard; Lavielle, Marc; Moulines, Eric. Convergence of a stochastic approximation version of the EM algorithm. Ann. Statist. 27 (1999), no. 1, 94--128. doi:10.1214/aos/1018031103. https://projecteuclid.org/euclid.aos/1018031103


Export citation

References

  • Andradottir, S. (1995). A stochastic approximation algorithm with varying bounds. Oper. Res. 1037-1048.
    Mathematical Reviews (MathSciNet): MR99c:65119
    Zentralblatt MATH: 0852.90115
    Digital Object Identifier: doi:10.1287/opre.43.6.1037
    JSTOR: links.jstor.org
  • Brandiere, O. and Duflo, M. (1996). Les algorithmes stochastiques contournent-ils les pi eges ? Ann. Inst. H. Poincar´e 32 395-427.
  • Brocker, T. (1975). Differentiable Germs and Catastrophes. Cambridge Univ. Press.
    Mathematical Reviews (MathSciNet): MR58:13132
  • Celeux, G. and Diebolt, J. (1988). A probabilistic teacher algorithm for iterative maximum likelihood estimation. In Classification and Related Methods of Data Analysis (H. H. Bock, ed.) 617-623. North-Holland, Amsterdam.
    Mathematical Reviews (MathSciNet): MR999548
  • Celeux, G. and Diebolt, J. (1992). A stochastic aprroximation type EM algorithm for the mixture problem. Stochastics Stochastics Rep. 41 127-146.
    Mathematical Reviews (MathSciNet): MR1275369
    Zentralblatt MATH: 0766.62050
  • Chen, H. F., Guo, L. and Gao, A. J. (1988). Convergence and robustness of the Robbins-Monro algorithm truncated at randomly varying bounds. Stoch. Process Appl. 27 217-231.
    Mathematical Reviews (MathSciNet): MR89b:62180
    Zentralblatt MATH: 0632.62082
    Digital Object Identifier: doi:10.1016/0304-4149(87)90039-1
  • Chickin, D. O. and Poznyak, A. S. (1984). On the asymptotical properties of the stochastic approximation procedure under dependent noise. Automat. Remote Control 44.
  • Chickin, D. O. (1988). On the convergence of the stochastic approximation procedure under dependent noise. Automat. Remote Control 48.
  • de Jong, P. and Shephard, N. (1995). The simulation smoother for time series model. Biometrika 82 339-350.
    Mathematical Reviews (MathSciNet): MR96j:62168
    Zentralblatt MATH: 0823.62072
    Digital Object Identifier: doi:10.1093/biomet/82.2.339
    JSTOR: links.jstor.org
  • Delyon, B. (1996). General results on stochastic approximation. IEEE Trans. Automat. Control. To appear.
    Mathematical Reviews (MathSciNet): MR1409470
    Digital Object Identifier: doi:10.1109/9.536495
  • Delyon, B. and Juditski, A. (1992). Stochastic approximation with averaging of trajectories. Stochastics Stochastics Rep. 39 107-118.
    Mathematical Reviews (MathSciNet): MR1275360
    Zentralblatt MATH: 0765.93081
  • Doukhan, P. (1994). Mixing: Properties and Examples. Lecture Notes in Statist. Springer, Berlin.
    Mathematical Reviews (MathSciNet): MR1312160
  • Diebolt, J. and Celeux, G. (1996). Asymptotic properties of a stochastic EM algorithm for estimating mixture proportions. Stochastic Models 9 599-613.
    Mathematical Reviews (MathSciNet): MR1249140
    Zentralblatt MATH: 0783.62021
    Digital Object Identifier: doi:10.1080/15326349308807283
  • Diebolt, J. and Ip, E. H. S. (1996). A stochastic EM algorithm for approximating the maximum likelihood estimate. In Markov Chain Monte Carlo in Practice (W. R. Gilks, S. T. Richardson, D. J. Spiegelhalter, eds.). Chapman and Hall, London.
    Mathematical Reviews (MathSciNet): MR1397971
    Zentralblatt MATH: 0840.62025
  • Duflo, M. (1997). Random Iterative Models. Springer, Berlin.
    Mathematical Reviews (MathSciNet): MR98m:62239
    Zentralblatt MATH: 0868.62069
  • Dempster, A., Laird, N. and Rubin, D. (1977). Maximum-likelihood from incomplete data via the EM algorithm. J. Roy. Statist. Soc. Ser. B 39 1-38.
    Mathematical Reviews (MathSciNet): MR58:18858
    Zentralblatt MATH: 0364.62022
    JSTOR: links.jstor.org
  • Fort, J. C. and Pag es, G. (1996). Convergence of stochastic algorithms: from Kushner-Clark theorem to the Lyapounov functional method. Adv. in Appl. Probab. 28 1072-1094.
    Mathematical Reviews (MathSciNet): MR1418247
    Zentralblatt MATH: 0881.62085
    Digital Object Identifier: doi:10.2307/1428165
    JSTOR: links.jstor.org
  • Geyer, C. J. and Thompson, E. A. (1992). Constrained Monte-Carlo maximum likelihood for dependent data. J. Roy. Statist. Soc. Ser. B 54 657-699.
    Mathematical Reviews (MathSciNet): MR94f:62042
    JSTOR: links.jstor.org
  • Geyer, C. J. (1994). On the convergence of Monte-Carlo maximum likelihoods calculations J. Roy. Statist. Soc. Ser. B 56 261-274.
    Mathematical Reviews (MathSciNet): MR1257812
    JSTOR: links.jstor.org
  • Geyer, C. J. (1996). Likelihood inference for spatial point processes. In Current Trends in Stochastic Geometry and its Applications (W. S. Kendall, O. E. Barndorff-Nielsen and M. C. van Lieshout, eds.) Chapman and Hall, London. To appear.
    Mathematical Reviews (MathSciNet): MR1673118
    Zentralblatt MATH: 0809.62089
  • Hall, P. and Heyde, C. C. (1980). Limit Theory and Its Applications. Academic Press, New York.
    Mathematical Reviews (MathSciNet): MR83a:60001
    Zentralblatt MATH: 0462.60045
  • Horn, R. and Johnson, C. (1985). Matrix Analysis. Cambridge Univ. Press.
    Mathematical Reviews (MathSciNet): MR87e:15001
  • Ibragimov, I. and Has'minski, R. (1981). Statistical Estimation: Asymptotic Theory. Springer, New York.
    Mathematical Reviews (MathSciNet): MR620321
  • Kushner, H. and Clark, D. (1978). Stochastic Approximation for Constrained and Unconstrained Systems. Springer, New York.
    Mathematical Reviews (MathSciNet): MR499560
  • Kushner, H. and Yin, G. (1997). Stochastic Approximation Algorithms and Applications. Springer, Berlin.
    Mathematical Reviews (MathSciNet): MR98h:62125
  • Lange, K. (1995). A gradient algorithm locally equivalent to the EM algorithm. J. Roy. Statist. Soc. Ser. B 75 425-437.
    Mathematical Reviews (MathSciNet): MR95k:62059
    Zentralblatt MATH: 0813.62021
    JSTOR: links.jstor.org
  • Liu, C. and Rubin, D. (1994). The ECME algorithm: a simple extension of EM and ECM with faster monotone convergence. Biometrika 81 633-648.
    Mathematical Reviews (MathSciNet): MR96f:62047
    Zentralblatt MATH: 0812.62028
    Digital Object Identifier: doi:10.1093/biomet/81.4.633
    JSTOR: links.jstor.org
  • Little, R. J. and Rubin, D. B. (1987). Statistical Analysis with Missing Data. Wiley, New York.
    Mathematical Reviews (MathSciNet): MR88i:62042
  • Louis, T. A. (1982). Finding the observed information matrix when using the EM algorithm. J. Roy. Statist. Soc. Ser. B 44 226-233.
    Mathematical Reviews (MathSciNet): MR83m:62004
    Zentralblatt MATH: 0488.62018
    JSTOR: links.jstor.org
  • MacDonald, I. L. and Zucchini, W. (1997). Hidden Markov and Other Models for Discrete-valued Time-series. Chapman and Hall, London.
    Mathematical Reviews (MathSciNet): MR2000g:62009
    Zentralblatt MATH: 0868.60036
  • Meilijson, I. (1989). A fast improvement to the EM algorithm on its own terms. J. Roy. Statist. Soc. Ser. B 51 127-138.
    Mathematical Reviews (MathSciNet): MR90a:62087
    Zentralblatt MATH: 0674.65118
    JSTOR: links.jstor.org
  • Meng, X. and Rubin, D. (1993). Maximum likelihood via the ECM algorithm: a general framework. Biometrika 80 267-278.
    Mathematical Reviews (MathSciNet): MR1243503
    Zentralblatt MATH: 0778.62022
    Digital Object Identifier: doi:10.1093/biomet/80.2.267
    JSTOR: links.jstor.org
  • Meng, X. (1994). On the rate of convergence of the ECM algorithm. Ann. Statist. 22 326-339.
    Mathematical Reviews (MathSciNet): MR95k:62060
    Zentralblatt MATH: 0803.65146
    Digital Object Identifier: doi:10.1214/aos/1176325371
    Project Euclid: euclid.aos/1176325371
  • Murray, G. (1977). Discussion on P. Dempster et al., Maximum-likelihood from incomplete data via the EM algorithm. J. Roy. Statist. Soc. 27-28.
    Mathematical Reviews (MathSciNet): MR501537
    JSTOR: links.jstor.org
  • Polyak, B. T. (1990). New stochastic approximation type procedures. Automatica i Telemekh. 98-107. (English translation in Automat. Remote Control 51.)
    Mathematical Reviews (MathSciNet): MR1071220
  • Polyak, B. T. and Juditski, A. B. (1992). Acceleration of stochastic approximation by averaging. SIAM J. Control Optim. 30 838-855.
    Mathematical Reviews (MathSciNet): MR93g:62110
    Zentralblatt MATH: 0762.62022
    Digital Object Identifier: doi:10.1137/0330046
  • Rao, C. (1965). Linear Statistical Inference and Its Applications. Wiley, New York.
    Mathematical Reviews (MathSciNet): MR36:4668
  • Shephard, N. (1994). Partial non-Gaussian state space. Biometrika 81 115-131.
    Mathematical Reviews (MathSciNet): MR95a:62073
    Zentralblatt MATH: 0796.62079
    Digital Object Identifier: doi:10.1093/biomet/81.1.115
    JSTOR: links.jstor.org
  • Tanner, M. (1993). Tools for Statistical Inference: Methods for Exploration of Posterior Distributions and Likelihood Functions. Springer, Berlin.
    Mathematical Reviews (MathSciNet): MR1238943
  • Titterington, D. M., Smith, A. F. M. and Makov, U. E. (1985). Statistical Analysis of Finite Mixture Distribution. Wiley, New York.
    Mathematical Reviews (MathSciNet): MR87j:62033
    Zentralblatt MATH: 0646.62013
  • Wei, G. and Tanner, M. (1990). A Monte-Carlo implementation of the EM algorithm and the Poor's Man's data augmentation algorithm. J. Amer. Statist. Assoc. 85 699-704.
  • Wu, C. (1983). On the convergence property of the EM algorithm. Ann. Statist. 11 95-103.
    Mathematical Reviews (MathSciNet): MR85e:62049
    Zentralblatt MATH: 0517.62035
    Digital Object Identifier: doi:10.1214/aos/1176346060
    Project Euclid: euclid.aos/1176346060
  • Younes, L. (1989). Parameter estimation for imperfectly observed Gibbsian fields. Probab. Theory and Related Fields 82 625-645.
    Mathematical Reviews (MathSciNet): MR1002904
    Zentralblatt MATH: 0659.62115
    Digital Object Identifier: doi:10.1007/BF00341287
  • Younes, L. (1992). Parameter estimation for imperfectly observed Gibbs fields and some comments on Chalmond's EM Gibbsian algorithm. Stochastic Models, Statistical Methods and Algorithms in Image Analysis. Lecture Notes in Statistics 74. Springer, Berlin.

The Institute of Mathematical Statistics

Annals of Statistics

New content alerts

Email RSS ToC RSS Article
  • You have access to this content.
  • You have partial access to this content.
  • You do not have access to this content.
Turn Off MathJax

What is MathJax?

More like this

+ See more