(This article was originally published at R – Xi'an's Og, and syndicated at StatsBlogs.)
Art Owen has arXived a new version of his thinning MCMC paper, where he studies how thinning or subsampling can improve computing time in MCMC chains. I remember quite well the message set by Mark Berliner and Steve MacEachern in an early 1990’s paper that subsampling was always increasing the variance of the resulting estimators. We actually have this result in our Monte Carlo Statistical Methods book. Now, there are other perspectives on this, as for instance cases when thinning can be hard-wired by simulating directly a k-step move, delaying rejection or acceptance, prefetching, or simulating directly the accepted values as in our vanilla Rao-Blackwellisation approach. Here, Art considers the case when there is a cost θ of computing a transform of the simulation [when the transition cost a unit] and when those transforms are positively correlated with correlation ρ. Somewhat unsurprisingly, when θ is large enough, thinning becomes worth implementing. But requires extra computations in evaluating the correlation ρ and the cost θ, which is rarely comparable with the cost of computing the likelihood itself, a requirement for the Metropolis-Hastings or Hamiltonian Monte Carlo step(s). Subsampling while keeping the right target (which is a hard constraint!) should thus have a much more effective impact on computing budgets.
Filed under: Books, pictures, R, Statistics Tagged: autocorrelation, computing time, MCMC, MCMC convergence, Monte Carlo Statistical Methods, thinning, vanilla Rao-Blackwellisation
Please comment on the article here: R – Xi'an's Og
