Statistical analysis of simulation output data
- 1 July 1976
- journal article
- Published by Association for Computing Machinery (ACM) in ACM SIGSIM Simulation Digest
- Vol. 7 (4) , 39-50
- https://doi.org/10.1145/1013610.807298
Abstract
This paper is a tutorial paper on how to obtain point estimates and confidence intervals of steady state means of simulation output data. The methods of using replications, batch means, and regenerative cycles for obtaining these point and interval estimates are discussed in detail and are applied to a simple time-shared computer model to illustrate their use. A brief discussion is included on using time series methods to obtain these estimates. The advantages and disadvantages of the various methods are given, including specific recommendations as to when certain methods might be used.Keywords
This publication has 8 references indexed in Scilit:
- Simulating Stable Stochastic Systems, VI: Quantile EstimationJournal of the ACM, 1976
- Introduction to Regenerative SimulationIBM Journal of Research and Development, 1975
- Investigation into Scheduling for an Interactive Computing SystemIBM Journal of Research and Development, 1974
- Simulating Stable Stochastic Systems, II: Markov ChainsJournal of the ACM, 1974
- Simulating Stable Stochastic Systems, I: General Multiserver QueuesJournal of the ACM, 1974
- A Time-Sharing Queue with a Finite Number of CustomersJournal of the ACM, 1969
- The Analysis of Simulation-Generated Time SeriesManagement Science, 1967
- Some Tactical Problems in Digital SimulationManagement Science, 1963