Optimal scaling for the pseudo-marginal random walk Metropolis: insensitivity to the noise generating mechanism
arXiv:1408.4344
Abstract
We examine the optimal scaling and the efficiency of the pseudo-marginal random walk Metropolis algorithm using a recently-derived result on the limiting efficiency as the dimension, . We prove that the optimal scaling for a given target varies by less than across a wide range of distributions for the noise in the estimate of the target, and that any scaling that is within of the optimal one will be at least efficient. We demonstrate that this phenomenon occurs even outside the range of distributions for which we rigorously prove it. We then conduct a simulation study on an example with where importance sampling is used to estimate the target density; we also examine results available from an existing simulations study with and where a particle filter was used. Our key conclusions are found to hold in these examples also.
arXiv commentNew version: simulation study now on a real statistical example (confusing typos corrected)
References in corpus (4)
- Renewal theory and computable convergence rates for geometrically ergodic Markov chains
- On the efficiency of pseudo-marginal random walk Metropolis algorithms
- On automating Markov chain Monte Carlo for a class of spatial models
- A Lognormal Central Limit Theorem for Particle Approximations of Normalizing Constants