Abstract
We discuss ASAP3, a refinement of the batch means algorithms ASAP and ASAP2. ASAP3 is a sequential procedure designed to produce a confidence-interval estimator for the expected response of a steady-state simulation that satisfies user-specified precision and coverage-probability requirements. ASAP3 operates as follows: the batch size is increased until the batch means pass the Shapiro-Wilk test for multivariate normality; and then ASAP3 fits a first-order autoregressive (AR(1)) time series model to the batch means. If necessary, the batch size is further increased until the autoregressive parameter in the AR(1) model does not significantly exceed 0.8. Next ASAP3 computes the terms of an inverse Cornish-Fisher expansion for the classical batch means t-ratio based on the AR(1) parameter estimates; and finally ASAP3 delivers a correlation-adjusted confidence interval based on this expansion. ASAP3 compared favorably with other batch means procedures (namely, ABATCH, ASAP, ASAP2, and LBATCH) in an extensive experimental performance evaluation.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have