Advertisement

On Implementation of the Markov Chain Monte Carlo Stochastic Approximation Algorithm

  • Yihua Jiang
  • Peter Karcher
  • Yuedong Wang
Chapter

Abstract

The Markov Chain Monte Carlo Stochastic Approximation Algorithm (MCMCSAA) was developed to compute estimates of parameters with incomplete data. In theory this algorithm guarantees convergence to the expected fixed points. However, due to its flexibility and complexity, care needs to be taken for implementation in practice. In this paper we show that the performance of MCMCSAA depends on many factors such as the Markov chain Monte Carlo sample size, the step-size of the parameter update, the initial values and the choice of an approximation to the Hessian matrix. Good choices of these factors are crucial to the practical performance and our results provide practical guidelines for these choices. We propose a new adaptive and hybrid procedure which is stable and faster while maintaining the same theoretical properties.

Keywords

Markov Chain Monte Carlo Monte Carlo Generalize Linear Mixed Model Hybrid Algorithm Stat Assoc 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Benveniste A, Métivier M, Priouret B (1987) Adaptive algorithms and stochastic approximations: Theory and applications of signal detection and processing and pattern recognition (French). Masson, ParisGoogle Scholar
  2. 2.
    Booth J, Hobert J (1999) Maximizing generalized linear mixed model likelihoods with an automated Monte Carlo em algorithm. J R Stat Soc Series B 61:265–285MATHCrossRefGoogle Scholar
  3. 3.
    Breslow NE, Clayton DG (1993) Approximate inference in generalized linear mixed models. J Am Stat Assoc 88:9–25MATHCrossRefGoogle Scholar
  4. 4.
    Gilks WR, Richardson S, Spiegelhalter DJE (1996) Markov chain Monte Carlo in practice. Chapman and Hall, LondonMATHGoogle Scholar
  5. 5.
    Gu MG, Kong FH (1998) A stochastic approximation algorithm with Markov chain Monte–Carlo for imcomplete data estimation problems. Proc Natl Acad Sci 95:7270–7274MATHCrossRefMathSciNetGoogle Scholar
  6. 6.
    Gu MG, Zhu HT (2001) Maximum likelihood estimation for spatial models by Markov chain Monte–Carlo stochastic approximation. J R Stat Soc Series B 63:339–355MATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    Jiang J (1998) Consistent estimators in generalized linear mixed models. J Am Stat Assoc 93:720–729MATHCrossRefGoogle Scholar
  8. 8.
    Karcher P, Wang Y (2002) Generalized nonparametric mixed effects models. J Comput Graph Stat 10:641–655MathSciNetGoogle Scholar
  9. 9.
    Kushner HJ, Yin GG (1997) Stochastic approximations algorithms and applications. Springer, New YorkGoogle Scholar
  10. 10.
    Lai TL (2003) Stochastic approximation. Ann Statis 31:391–406MATHCrossRefGoogle Scholar
  11. 11.
    Lin X, Breslow NE (1996) Bias correction in generalized linear mixed models with multiple components of dispersion. J Am Stat Assoc 91:1007–1016MATHCrossRefMathSciNetGoogle Scholar
  12. 12.
    Louis TA (1982) Finding the observed information matrix when using the em algorithm. J R Stat Soc Series B 44:226–233MATHMathSciNetGoogle Scholar
  13. 13.
    McCulloch C (1997) Maximum likelihood algorithms for generalized linear mixed models. J Am Stat Assoc 92:162–170MATHCrossRefMathSciNetGoogle Scholar
  14. 14.
    Satten GA (1996) Rank-based inference in the proportional hazards model for interval censored data. Biometrika 83:355–370MATHCrossRefGoogle Scholar
  15. 15.
    Zhu HT, Lee SY (2003) Analysis of generalized linear mixed models via a stochastic approximation algorithm with markov chain Monte–Carlo method. Stat Comput 391–406Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  1. 1.Capital One Financial Corp., 15000 Capital One Dr.RichmondUSA

Personalised recommendations