Batching methods in simulation output analysis: what we know and what we don’t

As an advanced tutorial, we discuss batching meth-ods for determining point-estimator precision for steady-state simulation experiments. We empha-size batching methods in which each batch pro-vides a point estimator analogous to that of the ex-periment, but we mention other methods that use batches, especially the more-general idea of stan-dardized time series. Despite the preponderance of literature on confidence-interval estimation for the mean using adjacent nonoverlapping batches, we fo-cus on estimating the point estimator’s standard er-ror and consider both general point estimators and general batching relationships. Literature on multi-variate batching exists, but we focus on the univari-ate problem. We consider the initial-transient prob-lem only in passing. Specific issues include form of the point estimator, definition of the batch st atis-tics, form of the batch-statistics estimator, optimal batch size (including various definitions of optimal), and determining batch size. This paper is a short summary of the issues, with a fairly complete bib-liography