Optimal scaling of random walk Metropolis algorithms using Bayesian large-sample asymptotics

Abstract

High-dimensional limit theorems have been shown useful to derive tuning rules for finding the optimal scaling in random walk Metropolis algorithms. The assumptions under which weak convergence results are proved are, however, restrictive: the target density is typically assumed to be of a product form. Users may thus doubt the validity of such tuning rules in practical applications. In this paper, we shed some light on optimal scaling problems from a different perspective, namely a large-sample one. This allows to prove weak convergence results under realistic assumptions and to propose novel parameter-dimension-dependent tuning guidelines. The proposed guidelines are consistent with the previous ones when the target density is close to having a product form, and the results highlight that the correlation structure has to be accounted for to avoid performance deterioration if that is not the case, while justifying the use of a natural (asymptotically exact) approximation to the correlation matrix that can be employed for the very first algorithm run.

Optimal scaling of MCMC beyond Metropolis

“…For MH algorithms, [3,34] relax the independence assumption, while [29] relax the identically distributed assumption. Additionally, [40] present a proof of weak convergence for MH for more general targets, and [33] provide optimal scaling results for general Bayesian targets using large-sample asymptotics. In these situations, extensions to other acceptance probabilities are similarly possible.…”