Advertisement

Methodology and Computing in Applied Probability

, Volume 13, Issue 3, pp 583–601 | Cite as

Optimal Scaling of Random Walk Metropolis Algorithms with Non-Gaussian Proposals

  • Peter Neal
  • Gareth Roberts
Article

Abstract

The asymptotic optimal scaling of random walk Metropolis (RWM) algorithms with Gaussian proposal distributions is well understood for certain specific classes of target distributions. These asymptotic results easily extend to any light tailed proposal distribution with finite fourth moment. However, heavy tailed proposal distributions such as the Cauchy distribution are known to have a number of desirable properties, and in many situations are preferable to light tailed proposal distributions. Therefore we consider the question of scaling for Cauchy distributed proposals for a wide range of independent and identically distributed (iid) product densities. The results are somewhat surprising as to when and when not Cauchy distributed proposals are preferable to Gaussian proposal distributions. This provides motivation for finding proposal distributions which improve on both Gaussian and Cauchy proposals, both for finite dimensional target distributions and asymptotically as the dimension of the target density, d → ∞. Throughout we seek the scaling of the proposal distribution which maximizes the expected squared jumping distance (ESJD).

Keywords

MCMC Cauchy distribution Spherical distributions Heavy tailed distributions  Random walk metropolis Optimal scaling 

AMS 2000 Subject Classifications

Primary 60F05; Secondary 65C05 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bédard M (2006) Weak convergence of Metropolis algorithms for non-IID target distributions. Ann Appl Probab 17:1222–1244CrossRefGoogle Scholar
  2. Breyer L, Roberts GO (2000) From Metropolis to diffusions: Gibbs states and optimal scaling. Stoch Process Appl 90:181–206MATHCrossRefMathSciNetGoogle Scholar
  3. Jarner S, Roberts G (2007) Convergence of heavy-tailed Monte Carlo Markov chain algorithms. Scand J Statist 34:781–815MATHMathSciNetGoogle Scholar
  4. Neal PJ, Roberts GO (2006) Optimal Scaling for partially updating MCMC algorithms. Ann Appl Probab 16:475–515MATHCrossRefMathSciNetGoogle Scholar
  5. Neal PJ, Roberts GO (2008) Optimal scaling for random walk metropolis on spherically constrained target densities. Method Comp Appl Prob 10:277–297MATHCrossRefMathSciNetGoogle Scholar
  6. Roberts GO, Gelman A, Gilks WR (1997) Weak convergence and optimal scaling of random walk metropolis algorithms. Ann Appl Probab 7:110–120MATHCrossRefMathSciNetGoogle Scholar
  7. Roberts GO, Rosenthal JS (2001) Optimal scaling for various metropolis-hastings algorithms. Stat Sci 16:351–367MATHCrossRefMathSciNetGoogle Scholar
  8. Sherlock C (2006) Methodology for inference on the Markov modulated Poisson process and theory for optimal scaling of the random walk Metropolis. PhD Thesis, Lancaster UniversityGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2010

Authors and Affiliations

  1. 1.School of MathematicsUniversity of ManchesterManchesterUK
  2. 2.Department of StatisticsUniversity of WarwickCoventryUK

Personalised recommendations