Summary
The choice of a suitable MCMC method and further the choice of a proposal distribution is known to be crucial for the convergence of the Markov chain. However, in many cases the choice of an effective proposal distribution is difficult. As a remedy we suggest a method called Adaptive Proposal (AP). Although the stationary distribution of the AP algorithm is slightly biased, it appears to provide an efficient tool for, e.g., reasonably low dimensional problems, as typically encountered in non-linear regression problems in natural sciences. As a realistic example we include a successful application of the AP algorithm in parameter estimation for the satellite instrument ‘GOMOS’. In this paper we also present systematic performance criteria for comparing Adaptive Proposal algorithm with more traditional Metropolis algorithms.
Similar content being viewed by others
References
Gelfand, A. E. & Smith, A. F. M. (1990), ‘Sampling-based approaches to calculate marginal densities’, J. Am. Stat. Ass. 85, 853–409.
Gelman, A. G., Roberts, G. O. & Gilks, W. R. (1996), Efficient Metropolis jumping rules, in J. M. Bernardo, J. O. Berger, A. F. David & A. F. M. Smith, eds, ‘Bayesian Statistics V’, Oxford University press, Oxford, pp. 599–608.
Gilks, W., Best, N. & Tan, K. (1995), ‘Adaptive rejection Metropolis sampling’, Applied Statistics 44, 455–472.
Gilks, W. R., Richardson, S. & Spiegelhalter, D. J. (1995), Introducing Markov chain Monte Carlo, in W. R. Gilks, S. Richardson & D. J. Spiegelhalter, eds, ‘Markov Chain Monte Carlo in Practice’, Chapman & Hall, pp. 1–19.
Gilks, W. R., Roberts, G. O. & George, E. I. (1994), ‘Adaptive direction sampling’, Statistical Science 43, 179–189.
Gilks, W. & Roberts, G. (1995), Stategies for improving MCMC, in W. R. Gilks, S. Richardson & D. J. Spiegelhalter, eds, ‘Markov Chain Monte Carlo in Practice’, Chapman & Hall, pp. 75–88.
Gilks, W., Roberts, G. & Sahu, S. (1998), ‘Adaptive Markov chain Monte Carlo through regeneration’, J. Amer. Statist. Assoc. To appear.
Gilks, W. & Wild, P. (1992), ‘Adaptive rejection sampling for Gibbs sampling’, Applied Statistics 41, 337–348.
Haario, H., Saksman, E. & Tamminen, J. (1998a), Adaptive proposal distribution for random walk Metropolis algorithm. Reports of the Department of Mathematics, University of Helsinki, Preprint 176.
Haario, H., Saksman, E. & Tamminen, J. (1998b), An Adaptive Metropolis algorithm. Reports of the Department of Mathematics, University of Helsinki, Preprint 187.
Hastings, W. (1970), ‘Monte Carlo sampling methods using Markov chains and their applications’, Biometrica 57, 97–109.
Kyrölä, E., Sihvola, E., Kotivuori, Y., Tikka, M., Tuomi, T. & Haario, H. (1993), ‘Inverse Theory for Occultation Measurements, 1. Spectral Inversion’, J. Geophys. Res. 98, 7367–7381.
Metropolis, N., Rosenbluth, A. W., Rosenbluth, M. N., Teller, A. H. & Teller, E. (1953), ‘Equations of state calculations by fast computing machine’, J. Chem. Phys. 21, 1087–1091.
Press, W. H., Teukolsky, S. A., Vetterling, W. T. & Flannery, B. P. (1992), Numerical Recipes in FORTRAN, The art of scientific computing, Clasendon Press, Oxford.
Roberts, G. O., Gelman, A. & Gilks, W. R. (1994), Weak convergence and optimal scaling of random walk metropolis algorithms. Preprint. URL: https://doi.org/www.stats.bris.ac.uk/MCMC/
Sahu, S. K. & Zhigljavsky, A. A. (1998a), Self regenerative Markov chain Monte Carlo. Preprint. URL: https://doi.org/www.stats.bris.ac.uk/MCMC/
Sahu, S. K. & Zhigljavsky, A. A. (1998b), Adaptation for self regenerative MCMC. Preprint. URL: https://doi.org/www.stats.bris.ac.uk/MCMC/
Acknowledgments
The second author (E.S) has been supported by the Academy of Finland, Project 32837. We thank Elja Arjas, Esa Nummelin and Antti Penttinen for useful discussions on the topics of the paper.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Haario, H., Saksman, E. & Tamminen, J. Adaptive proposal distribution for random walk Metropolis algorithm. Computational Statistics 14, 375–395 (1999). https://doi.org/10.1007/s001800050022
Published:
Issue Date:
DOI: https://doi.org/10.1007/s001800050022