Coupling a stochastic approximation version of EM with an MCMC procedure

The stochastic approximation version of EM (SAEM) proposed by Delyon et al. (1999) is a powerful alternative to EM when the E-step is intractable. Convergence of SAEM toward a maximum of the observed likelihood is established when the unobserved data are simulated at each iteration under the conditional distribution. We show that this very restrictive assumption can be weakened. Indeed, the results of Benveniste et al. for stochastic approximation with Markovian perturbations are used to establish the convergence of SAEM when it is coupled with a Markov chain Monte-Carlo procedure. This result is very useful for many practical applications. Applications to the convolution model and the change-points model are presented to illustrate the proposed method.

[1]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[2]  New York Dover,et al.  ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM , 1983 .

[3]  Han-Fu Chen,et al.  Convergence and robustness of the Robbins-Monro algorithm truncated at randomly varying bounds , 1987 .

[4]  Pierre Priouret,et al.  Adaptive Algorithms and Stochastic Approximations , 1990, Applications of Mathematics.

[5]  Xiao-Li Meng,et al.  Maximum likelihood estimation via the ECM algorithm: A general framework , 1993 .

[6]  Richard L. Tweedie,et al.  Markov Chains and Stochastic Stability , 1993, Communications and Control Engineering Series.

[7]  K. Lange.  A gradient algorithm locally equivalent to the EM algorithm , 1995 .

[8]  O. Brandière,et al.  Les algorithmes stochastiques contournent-ils les pièges? , 1995 .

[9]  R. Tweedie,et al.  Rates of convergence of the Hastings and Metropolis algorithms , 1996 .

[10]  Eric Moulines,et al.  A simulated annealing version of the EM algorithm for non-Gaussian deconvolution , 1997, Stat. Comput..

[11]  F. Kong,et al.  A stochastic approximation algorithm with Markov chain Monte-carlo method for incomplete data estimation problems. , 1998, Proceedings of the National Academy of Sciences of the United States of America.

[12]  É. Moulines,et al.  Convergence of a stochastic approximation version of the EM algorithm , 1999 .

[13]  Jianfeng Yao.  On Recursive Estimation in Incomplete Data Models , 2000 .

[14]  Marc Lavielle,et al.  An application of MCMC methods for the multiple change-points problem , 2001, Signal Process..

[15]  M. Gu,et al.  Maximum likelihood estimation for spatial models by Markov chain Monte Carlo stochastic approximation , 2001 .

[16]  Didier Concordet,et al.  A simulated pseudo-maximum likelihood estimator for nonlinear mixed models , 2002 .