Announcement_7

Paper accepted at Annals of Statistics:

The focus of the paper is the bias of importance sampling and independent Metropolis-Hastings, without assuming that the weight function is bounded, and how to remove that bias using couplings. We show that the common random numbers coupling of independent Metropolis-Hastings is maximal. Using that coupling, we derive polynomial bounds on the total variation distance of the chain to its target distribution and provide moment bounds for the proposed unbiased estimators.