A comparison between standardized time series and overlapping batched means

ABSTRACT

We study several methods for estimating the variance of the sample mean from a stationary stochastic process. Particular emphasis is placed on comparing the recently introduced methods of standardized time series and overlapping batched means. Comparison criteria include estimator bias and variance.

REFERENCES

• 1: Go}dsman, D. (1984). On Using Standardize~ T,;me Series to Analyze Stochastic Processes. Ph.D. Di ssertation, School of 0.12. & I.E. , Cornel I Lln~versity, Ithaca, NY.
• 2: Goldsman, 0. and M.S. Meketon (1985). "A Compari son of Several Vari ance Estimators". Tech. Report, School of ISyE, Georgia Institute of Technology, Atlanta, GA.
• 3: Goldsman, O. and L. Schruben (1984). "Asymptotic Properties of Some Confidence Interval Estimators for Simulation Output". Management Science~ 30, 1217--1225i.
• 4: Law, A.M. and W.D. Kel ton (1982). "Confldence Interval s for Steady-State S i mul a t i ons, I : A Survey o f F i xed Samp I e Si ze Procedures". Tech. Report #82--5, Dept. of Management Information Systems, Universi ty of Ari zona, Tucson, AZ.
• 5: Meketon, M.S. (1980). The Variance Time Curve: Theory, Estimation and Application. Ph.O. Di,~sertation, School of O.R. & I.E., Cornel I Uni versi ty, Ithaoa, NY.
• 6: Marc S. Meketon , Bruce Schmeiser, Overlapping batch means: something for nothing?, Proceedings of the 16th conference on Winter simulation, p.226-230, December 28-30, 1984, Dallas, TX
• 7: Schruben, L. (1983)."'Confidence Interval Estimation Usin9 Standardized Time Ser i e s ". Opera t ions l~esearch, 311 , 1090--i 108.