Advertisement

Statistics and Computing

, Volume 6, Issue 2, pp 113–119 | Cite as

Metropolized independent sampling with comparisons to rejection sampling and importance sampling

  • Jun S. Liu
Papers

Abstract

Although Markov chain Monte Carlo methods have been widely used in many disciplines, exact eigen analysis for such generated chains has been rare. In this paper, a special Metropolis-Hastings algorithm, Metropolized independent sampling, proposed first in Hastings (1970), is studied in full detail. The eigenvalues and eigenvectors of the corresponding Markov chain, as well as a sharp bound for the total variation distance between the nth updated distribution and the target distribution, are provided. Furthermore, the relationship between this scheme, rejection sampling, and importance sampling are studied with emphasis on their relative efficiencies. It is shown that Metropolized independent sampling is superior to rejection sampling in two respects: asymptotic efficiency and ease of computation.

Keywords

Coupling delta method eigen analysis importance ratio 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Besag, J. and Green, P.J. (1993). Spatial statistics and Baysian computation. Journal of the Royal Statistical Society B, 55, 24–35.Google Scholar
  2. Diaconis, P. (1988) Group Representations in Probability and Statistics, Lecture Notes-Monograph Series 11, IMS, Hayward California.Google Scholar
  3. Diaconis, P. and Hanlon, P. (1992) Eigen analysis for some examples of the Metropolis algorithm. Contemporary Mathematics, 138, 99–117.Google Scholar
  4. Gelfand, A. E. and Smith, A. F. M. (1990) Sampling-based approaches to calculating marginal densities. Journal of the American Statistical Association, 85, 398–409.Google Scholar
  5. Gelman, A. and Rubin, D. B. (1993) Discussion on Gibbs sampler and other MCMC methods. Journal of the Royal Statistical Society B, 55, 73–73.Google Scholar
  6. Geman, S. and Geman, D. (1984) Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 6, 721–41.Google Scholar
  7. Hastings, W. K. (1970) Monte Carlo sampling methods using Markov chains and their applications. Biometrika, 57, 97–109.Google Scholar
  8. Kong, A. (1992) A note on importance sampling using renormalized weights. Technical report, Department of Statistics, University of Chicago.Google Scholar
  9. Lindvall, T. (1992) Lectures on the Coupling Method. Wiley, New York.Google Scholar
  10. Liu, J. S. (1994) The collapsed Gibbs sampler in Bayesian computations with applications to a gene regulation problem. Joural of the American Statistical Association, 89, 958–66.Google Scholar
  11. Liu, J. S., Kong, A. and Wong, W. H. (1994) Covariance structure of the Gibbs sampler with applications to the comparisons of estimators and augmentation schemes. Biometrika, 81, 27–40.Google Scholar
  12. Lovasz, L. and Simonovits, M. (1990) The mixing rate of Markov chains, an isoperimetric inequality, and computing the volume. Preprint 27, Hungarian Academy of Sciences.Google Scholar
  13. Marshall, A. W. (1956) The use of multi-stage sampling schemes in Monte Carlo computations. In Symposium on Monte Carlo Methods, ed. M. A. Meyer, pp. 123–40, Wiley, New York.Google Scholar
  14. Metropolis, N., Rosenbluth, A. W., Rosenbluth, M. N., Teller, A. H. and Teller, E. (1953) Equations of state calculations by fast computing machines. Journal of Chemical Physics, 21, 1087–91.Google Scholar
  15. von Neumann, J. (1951) Various techniques used in connection with random digits. National Bureau of Standards Applied Mathematics Series, 12, 36–8.Google Scholar
  16. Peskun, P. H. (1973) Optimum Monte-Carlo sampling using Markov chains. Biometrika, 60, 607–12.Google Scholar
  17. Rosenthal, J. S. (1995) Rates of convergence for Gibbs sampler for variance components models. Ann. Statist., 23, 740–61.Google Scholar
  18. Smith, A. F. M. and Roberts, G. O. (1993) Bayesian Computation via the Gibbs sampler and related Markov chain Monte Carlo methods. Journal of the Royal Statistical Society, B, 55, 3–23.Google Scholar
  19. Smith, R. L. (1994) Exact transition probabilities for Metropolized independent sampling. Technical Report, Dept. Statistics, Univ. of North Carolina.Google Scholar
  20. Tanner, M. A. and Wong, W. H. (1987) The calculation of posterior distributions by data augmentation (with discussion). Journal of American Statistical Association, 82, 528–50.Google Scholar
  21. Tierney, L. (1991) Markov chains for exploring posterior distributions. In Computer Science and Statistics: Proc. 23rd Symp. Interface.Google Scholar
  22. Yoida, K. (1978) Functional Analysis. Springer-Verlag, New York.Google Scholar

Copyright information

© Chapman & Hall 1996

Authors and Affiliations

  • Jun S. Liu
    • 1
  1. 1.Department of StatisticsStanford UniversityStanfordUSA

Personalised recommendations