Overlapping variance estimators for simulations

We examine properties of overlapped versions of the standardized time series area and Cramer-von Mises estimators for the variance parameter of a stationary stochastic process, e.g., a steady-state simulation output process. We find that the overlapping estimators have the same bias properties as, but lower variance than, their nonoverlapping counterparts; the new estimators also perform well against the benchmark batch means estimator. We illustrate our findings with analytical and Monte Carlo examples.