Standardized time series methods: variance estimation using replicated batch means

ABSTRACT

We present a new method for obtaining confidence intervals in steady-state simulation. In our replicated batch means method, we do a small number of independent replications to estimate the steady-state mean of the underlying stochastic process. In order to obtain a variance estimator, we further group the observations from these replications into non-overlapping batches. We show that for large sample sizes, the new variance estimator is less biased than the batch means variance estimator, the variances of the two variance estimators are approximately equal, and the new steady-state mean estimator has a smaller variance than the batch means estimator when there is positive serial correlation between the observations. For small sample sizes, we compare our replicated batch means method with the (standard) batch means and multiple replications methods empirically, and show that the best overall coverage of confidence intervals is obtained by the replicated batch means method with a small number of replications.

References

1. Alexopoulos, C., and A. F. Seila. 1998. Output Data Analysis. Chapter 7 in the Handbook of Simulation---Principles, Methodology, Advances, Applications, and Practice, ed. J. Banks, 225-272. New York: John Wiley and Sons, Inc.
2. Andradóttir, S., and N. T. Argon. 2001. Replicated batch means for steady-state simulation. Working paper.
3. Chien, C. H., D. Goldsman, and B. Melamed. 1997. Large-sample results for batch means. Management Science 43 (9): 1288-1295.
4. Glynn, P. W., and D. L. Iglehart. 1990. Simulation output analysis using standardized time series. Mathematics of Operations Research 15 (1): 1-16.
5. Kelton, W. D., and A. M. Law. 1984. An analytical evaluation of alternative strategies in steady-state simulation. Operations Research 32 (1): 169-184.
6. Lewis, P. A. W. 1980. Simple models for positive-valued and discrete-valued time series with ARMA correlation structure. In Multivariate Analysis---V: Proceedings of the Fifth International Symposium on Multivariate Analysis, ed. P. R. Krishnaiah, 151-166. New York: North-Holland.
7. Song, W.-M. T., and B. W. Schmeiser. 1995. Optimal mean-squared-error batch sizes. Management Science 41 (1): 110-123.
8. Whitt, W. 1991. The efficiency of one long run versus independent replications in steady-state simulation. Management Science 37 (6): 645-666.

Index Terms

• Mathematics of computing

  • Probability and statistics

    • Statistical paradigms

      • Statistical graphics

      • Time series analysis

    • Stochastic processes