Does Waste-recycling Really Improve the Multi-proposal Metropolis-hastings Monte Carlo Algorithm?

The waste-recycling Monte Carlo (WR) algorithm introduced by physicists is a modification of the (multi-proposal) Metropolis-Hastings algorithm, which makes use of all the proposals in the empirical mean, whereas the standard (multi-proposal) Metropolis-Hastings algorithm only uses the accepted proposals. In this paper, we extend the WR algorithm into a general control variate technique and exhibit the optimal choice of the control variate in terms of asymptotic variance. We also give an example which shows that in contradiction to the intuition of physicists, the WR algorithm can have an asymp-totic variance larger than the one of the Metropolis-Hastings algorithm. However, in the particular case of the Metropolis-Hastings algorithm called Boltzmann algorithm, we prove that the WR algorithm is asymptotically better than the Metropolis-Hastings algorithm. This last property is also true for the multi-proposal Metropolis-Hastings algorithm. In this last framework, we consider a linear parametric generalization of WR, and we propose an estimator of the explicit optimal parameter using the proposals.