2008/11/30 by James M. Flegal, Galin L. Jones · 1 citation
Mathematics · #math.ST #stat.TH #msc:60J22 #msc:62M15
paper · pdf · doi:10.1214/09-aos735
published as Annals of Statistics 2010, Vol. 38, No. 2, 1034-1070 · Published in at http://dx.doi.org/10.1214/09-AOS735 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
arxiv created 2010/02/25 · arxiv updated 2010/02/26
Calculating a Monte Carlo standard error (MCSE) is an important step in the statistical analysis of the simulation output obtained from a Markov chain Monte Carlo experiment. An MCSE is usually based on an estimate of the variance of the asymptotic normal distribution. We consider spectral and batch means methods for estimating this variance. In particular, we establish conditions which guarantee that these estimators are strongly consistent as the simulation effort increases. In addition, for the batch means and overlapping batch means methods we establish conditions ensuring consistency in the mean-square sense which in turn allows us to calculate the optimal batch size up to a constant of proportionality. Finally, we examine the empirical finite-sample properties of spectral variance and batch means estimators and provide recommendations for practitioners.