Confidence intervals in discrete event simulation: A comparison of replication and batch means
Author:
- 1 December 1977
- journal article
- research article
- Published by Wiley in Naval Research Logistics Quarterly
- Vol. 24 (4) , 667-678
Abstract
Suppose that we have 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. We can make k independent runs of m observations each (n=k.m) or, alternatively, one run of n observations which we then divide 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, we empirically show that batch means is superior to replication, but that neither method works well if n is too small. We also show that if m is chosen too small for replication, then the coverage may decrease dramatically as the total sample size n is increased.Keywords
This publication has 6 references indexed in Scilit:
- Simulating stable stochastic systems, V: Comparison of ratio estimatorsNaval Research Logistics Quarterly, 1975
- Simulating Stable Stochastic Systems, II: Markov ChainsJournal of the ACM, 1974
- Simulating Stable Stochastic Systems, I: General Multiserver QueuesJournal of the ACM, 1974
- An Approximation for the Moments of Waiting TimesOperations Research, 1969
- The serial correlation coefficients of waiting times in a stationary single server queueJournal of the Australian Mathematical Society, 1968
- Fourier Analysis of Uniform Random Number GeneratorsJournal of the ACM, 1967