Confidence Intervals in Discrete Event Simulation: A Comparison of Replication and Batch Means.

摘要

Suppose that one has enough computer time to make n observations of a stochastic process by means of simulation and would like to construct a confidence interval for the steady-state mean. One can make k independent runs of m observations each (n=k.m) or, alternatively, one run of n observations which one then divides into k batches of length m. These methods are known as replication and batch means, respectively. In this paper, using the probability of coverage and the half length of a confidence interval as criteria for comparison, the author empirically shows that batch means is superior to replication, but that neither method works well the author n is too small. He also shows that if m is chosen too small for replication, then the coverage may decrease dramatically as the total sample size n is increased. (Author)