Paper No. 09-35
K Latuszynski and W Niemiro
Rigorous confidence bounds for MCMC under a geometric drift condition
Abstract: We assume a drift condition towards a small set and bound the mean square error of estimators obtained by taking averages along a single trajectory of a Markov chain Monte Carlo algorithm. We use these bounds to construct fixed-width nonasymptotic confidence intervals.
Keywords: MCMC estimation, confidence intervals, mean square error, Markov chain, convergence rate, V-uniform ergodicity, drift condition, simulation cost.