Abstract
Different strategies have been proposed to improve mixing and convergence properties of Markov Chain Monte Carlo algorithms. These are mainly concerned with customizing the proposal density in the Metropolis–Hastings algorithm to the specific target density and require a detailed exploratory analysis of the stationary distribution and/or some preliminary experiments to determine an efficient proposal. Various Metropolis–Hastings algorithms have been suggested that make use of previously sampled states in defining an adaptive proposal density. Here we propose a general class of adaptive Metropolis–Hastings algorithms based on Metropolis–Hastings-within-Gibbs sampling. For the case of a one-dimensional target distribution, we present two novel algorithms using mixtures of triangular and trapezoidal densities. These can also be seen as improved versions of the all-purpose adaptive rejection Metropolis sampling (ARMS) algorithm to sample from non-logconcave univariate densities. Using various different examples, we demonstrate their properties and efficiencies and point out their advantages over ARMS and other adaptive alternatives such as the Normal Kernel Coupler.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Besag, J.E.: Comment on ‘Representations of knowledge in complex systems’ by U. Grenander and M.I. Miller. J. Roy. Stat. Soc. Ser. B 56, 591–592 (1994)
Besag, J.E., Green, P.J., Higdon, D., Mengersen, K.: Bayesian computation and stochastic systems. With comments and a reply by the authors. Stat. Sci. 10, 3–66 (1995)
Brockwell, A.E., Kadane, J.B.: Identification of regeneration times in MCMC simulation, with application to adaptive schemes. J. Comput. Graph. Stat. 14, 436–458 (2005)
Carlin, B.P., Polson, N.G., Stoffer, D.S.: A Monte Carlo approach to nonnormal and nonlinear state-space modeling. J. Am. Stat. Assoc. 87, 493–500 (1992)
Chauveau, D., Vandekerkhove, P.: Improving convergence of the Hastings Metropolis algorithm with an adaptive proposal. Scand. J. Stat. 29, 13–29 (2002)
deBoor, C.: A Practical Guide to Splines. Springer, New York (1978)
Devroye, L.: Non-uniform Random Variate Generation. Springer, New York (1986)
Fahrmeir, L., Tutz, G.: Multivariate Statistical Modeling Based on Generalized Linear Models. Springer, New York (1994)
Gelfand, A.E., Sahu, S.K.: On Markov chain Monte Carlo acceleration. J. Am. Stat. Assoc. 3, 261–276 (1994)
Gelfand, A.E., Smith, A.F.M.: Sampling-based approaches to calculating marginal densities. J. Am. Stat. Assoc. 85, 398–409 (1990)
Geman, S., Geman, D.: Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images. IEEE Trans. Pattern Anal. Mach. Intell. 6, 721–741 (1984)
Geweke, J.: Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. In: Bernardo, J.M., Berger, J.O., Dawid, A.P., Smith, A.F.M. (eds.) Bayesian Statistics, vol. 4, pp. 169–193. Oxford University Press, Oxford (1992)
Gilks, W.R., Roberts, G.O.: Adaptive Markov Chain Monte Carlo. IMS Bull. 22, 275 (1993)
Gilks, W.R., Roberts, G.O.: Strategies for improving MCMC. In: Gilks, W.R., et al. (eds.) Markov Chain Monte Carlo in Practice. pp. 89–114. Chapman & Hall, London (1996)
Gilks, W.R., Roberts, G.O., George, E.I.: Adaptive direction sampling. Statistician, 43, 179–189 (1994)
Gilks, W.R., Best, N.G., Tan, K.K.C.: Adaptive rejection Metropolis sampling within Gibbs sampling. Appl. Stat. 44, 455–472 (1995)
Gilks, W.R., Richardson, S., Spiegelhalter, D.J.: Markov Chain Monte Carlo in Practice. Chapman & Hall, London (1996)
Gilks, W.R., Neal, R.M., Best, N.G., Tan, K.K.C.: Corrigendum: Adaptive rejection Metropolis sampling. Appl. Stat. 46, 541–542 (1997)
Gilks, W.R., Roberts, G.O., Sahu, S.K.: Adaptive Markov chain Monte Carlo through regeneration. J. Am. Stat. Assoc. 93, 1045–1054 (1998)
Haario, H., Saksman, E., Tamminen, J.: Adaptive proposal distribution for random walk Metropolis algorithm. Comput. Stat. 14, 375–395 (1999)
Haario, H., Saksman, E., Tamminen, J.: An adaptive Metropolis algorithm. Bernoulli 7, 223–242 (2001)
Harvey, A.: Forecasting, Structural Time Series Models and the Kalman Filter. Cambridge University Press, New York (1989)
Hastings, W.K.: Monte Carlo sampling methods using Markov chains. Biometrika 57, 97–109 (1970)
Hilborn, R., Walters, C.J.: Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty. Chapman & Hall, New York (1992)
Holden, L.: Adaptive chains. Technical report, Norwegian Computing Center (2000)
Kalman, R.E.: A new approach to linear filtering and prediction problems. J. Basic Eng. 82, 34–45 (1960)
Kuensch, H.R.: State space and hidden Markov models. In: Barndorff-Nielsen et al.. (eds.) Complex Stochastic Systems, pp. 109–174. Chapman & Hall, London (2001)
Liu, J.S., Sabatti, C.: Generalised Gibbs sampler and multigrid Monte Carlo for Bayesian computation. Biometrika 87, 353–369 (2000)
Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A.H., Teller, E.: Equations of state calculations by fast computing machines. J. Chem. Phys. 21, 1087–1091 (1953)
Meyer, R., Christensen, N.L.: Bayesian reconstruction of chaotic dynamical systems. Phys. Rev. E 62, 3535–3542 (2000)
Meyer, R., Millar, R.B.: Bayesian stock assessment using a state-space implementation of the delay difference model. Can. J. Fish. Aquat. Sci. 56, 37–52 (1999)
Meyer, R., Yu, J.: BUGS for a Bayesian analysis of stochastic volatility models. Econ. J. 3(2), 198–215 (2000)
Mira, A., Sargent, D.: A new strategy for speeding Markov chain Monte Carlo algorithms. Stat. Methods Appl. 1(12), 49–60 (2003)
Mykland, P., Tierney, L., Yu, B.: Regeneration in Markov chain samplers. J. Am. Stat. Assoc. 90, 233–241 (1995)
Pella, J.J., Tomlinson, P.K.: A generalized stock production model. Inter-Am. Trop. Tuna Commun. Bull. 13, 421–496 (1969)
Perron, F., Mengersen, K.: Bayesian nonparametric modelling using mixtures of triangular distributions. Biometrics 57(2), 518–529 (2001)
Roberts, G.O., Gilks, W.R.: Convergence of adaptive direction sampling. J. Multivar. Anal. 49, 287–298 (1994)
Roberts, G.O., Rosenthal, J.S.: Optimal scaling for various Metropolis–Hastings algorithms. Stat. Sci. 16, 351–367 (2001)
Roberts, G., Rosenthal, J.: Harris recurrence of Metropolis-within-Gibbs and trans-dimensional Markov chains. Ann. Appl. Probab. 16(4), 2123–2139 (2006a)
Roberts, G.O., Rosenthal, J.S.: Examples of adaptive MCMC. Preprint (2006b)
Roberts, G.O., Rosenthal, J.S.: Coupling and ergodicity of adaptive MCMC. J. Appl. Probab. 44, 458–475 (2007)
Roberts, G.O., Tweedie, R.L.: Exponential convergence of Langevin diffusions and their discrete approximations. Biometrika 2, 341–363 (1996)
Sahu, S.K., Zhigljavsky, A.A.: Self regenerative Markov chain Monte Carlo with adaptation. Bernoulli 9(3), 395–422 (2003)
Schumaker, L.L.: Spline Functions: Basic Theory. Krieger, Melbourne (1993)
Sokal, A.D.: Monte Carlo Methods in Statistical Mechanics: Foundations and New Algorithms. Lectures at the Cargese Summer School on “Functional Integration: Basics and Applications” (1996)
Tierney, L.: Markov chains for exploring posterior distributions. Ann. Stat. 22, 1701–1762 (1994)
Tierney, L., Mira, A.: Some adaptive Monte Carlo methods for Bayesian inference. Stat. Med. 18, 2507–2515 (1999)
Warnes, G.R.: The normal kernel coupler: An adaptive Markov chain Monte Carlo method for efficiently sampling from multi-modal distributions. Technical report No. 39, Department of Statistics, University of Washington (2001)
West, M., Harrison, P.J.: Bayesian Forecasting and Dynamic Models, 2nd edn. Springer, New York (1997)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cai, B., Meyer, R. & Perron, F. Metropolis–Hastings algorithms with adaptive proposals. Stat Comput 18, 421–433 (2008). https://doi.org/10.1007/s11222-008-9051-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-008-9051-5