The Annals of Applied Probability

Optimal scaling of random walk Metropolis algorithms with discontinuous target densities

Peter Neal, Gareth Roberts, and Wai Kong Yuen

Full-text: Open access

Abstract

We consider the optimal scaling problem for high-dimensional random walk Metropolis (RWM) algorithms where the target distribution has a discontinuous probability density function. Almost all previous analysis has focused upon continuous target densities. The main result is a weak convergence result as the dimensionality $d$ of the target densities converges to $\infty$. In particular, when the proposal variance is scaled by $d^{-2}$, the sequence of stochastic processes formed by the first component of each Markov chain converges to an appropriate Langevin diffusion process. Therefore optimizing the efficiency of the RWM algorithm is equivalent to maximizing the speed of the limiting diffusion. This leads to an asymptotic optimal acceptance rate of $e^{-2}$ $(=0.1353)$ under quite general conditions. The results have major practical implications for the implementation of RWM algorithms by highlighting the detrimental effect of choosing RWM algorithms over Metropolis-within-Gibbs algorithms.

Article information

Source
Ann. Appl. Probab., Volume 22, Number 5 (2012), 1880-1927.

Dates
First available in Project Euclid: 12 October 2012

Permanent link to this document
https://projecteuclid.org/euclid.aoap/1350067989

Digital Object Identifier
doi:10.1214/11-AAP817

Mathematical Reviews number (MathSciNet)
MR3025684

Zentralblatt MATH identifier
1259.60082

Subjects
Primary: 60F05: Central limit and other weak theorems
Secondary: 65C05: Monte Carlo methods

Keywords
Random walk Metropolis Markov chain Monte Carlo optimal scaling

Citation

Neal, Peter; Roberts, Gareth; Yuen, Wai Kong. Optimal scaling of random walk Metropolis algorithms with discontinuous target densities. Ann. Appl. Probab. 22 (2012), no. 5, 1880--1927. doi:10.1214/11-AAP817. https://projecteuclid.org/euclid.aoap/1350067989


Export citation

References

  • [1] Barbour, A. D. (1988). Stein’s method and Poisson process convergence. J. Appl. Probab. 25A 175–184.
  • [2] Barbour, A. D., Holst, L. and Janson, S. (1992). Poisson Approximation. Oxford Studies in Probability 2. Oxford Univ. Press, Oxford.
  • [3] Bédard, M. (2007). Weak convergence of Metropolis algorithms for non-i.i.d. target distributions. Ann. Appl. Probab. 17 1222–1244.
  • [4] Bédard, M. (2008). Optimal acceptance rates for Metropolis algorithms: Moving beyond 0.234. Stochastic Process. Appl. 118 2198–2222.
  • [5] Billingsley, P. (1968). Convergence of Probability Measures. Wiley, New York.
  • [6] Billingsley, P. (1979). Probability and Measure. Wiley, New York.
  • [7] Breyer, L. A. and Roberts, G. O. (2000). From Metropolis to diffusions: Gibbs states and optimal scaling. Stochastic Process. Appl. 90 181–206.
  • [8] Christensen, O. F., Roberts, G. O. and Rosenthal, J. S. (2005). Scaling limits for the transient phase of local Metropolis–Hastings algorithms. J. R. Stat. Soc. Ser. B Stat. Methodol. 67 253–268.
  • [9] Ethier, S. N. and Kurtz, T. G. (1986). Markov Processes, Characterization and Convergence. Wiley, New York.
  • [10] Neal, P. and Roberts, G. (2006). Optimal scaling for partially updating MCMC algorithms. Ann. Appl. Probab. 16 475–515.
  • [11] Neal, P. and Roberts, G. (2008). Optimal scaling for random walk Metropolis on spherically constrained target densities. Methodol. Comput. Appl. Probab. 10 277–297.
  • [12] Neal, P. J. and Roberts, G. O. (2011). Optimal scaling of random walk Metropolis algorithms with non-Gaussian proposals. Methodol. Comput. Appl. Probab. 13 583–601.
  • [13] Revuz, D. and Yor, M. (1998). Continuous Martingales and Brownian Motion. Springer, Berlin.
  • [14] Roberts, G. O., Gelman, A. and Gilks, W. R. (1997). Weak convergence and optimal scaling of random walk Metropolis algorithms. Ann. Appl. Probab. 7 110–120.
  • [15] Roberts, G. O. and Rosenthal, J. S. (2001). Optimal scaling for various Metropolis–Hastings algorithms. Statist. Sci. 16 351–367.
  • [16] Sherlock, C. (2006). Methodology for inference on the Markov modulated Poisson process and theory for optimal scaling of the random walk Metropolis. Ph.D. thesis, Lancaster Univ.
  • [17] Sherlock, C. and Roberts, G. (2009). Optimal scaling of the random walk Metropolis on elliptically symmetric unimodal targets. Bernoulli 15 774–798.