On the Asymptotic Validity of Fully Sequential Selection Procedures for Steady-State Simulation
Authors:
- 1 June 2006
- journal article
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 54 (3) , 475-488
Abstract
We present fully sequential procedures for steady-state simulation that are designed to select the best of a finite number of simulated systems when “best” is defined by the largest or smallest long-run average performance. We also provide a framework for establishing the asymptotic validity of such procedures and prove the validity of our procedures. An example based on the M/M/1 queue is given.Keywords
This publication has 23 references indexed in Scilit:
- Some Almost-Sure Convergence Properties Useful in Sequential AnalysisSequential Analysis, 2005
- A fully sequential procedure for indifference-zone selection in simulationACM Transactions on Modeling and Computer Simulation, 2001
- Computing efforts allocation for ordinal optimization and discrete event simulationIEEE Transactions on Automatic Control, 2000
- Two-stage multiple-comparison procedures for steady-state simulationsACM Transactions on Modeling and Computer Simulation, 1999
- Comparing Systems via Simulation1998
- Multiple-comparison procedures for steady-state simulationsThe Annals of Statistics, 1997
- The Asymptotic Validity of Sequential Stopping Rules for Stochastic SimulationsThe Annals of Applied Probability, 1992
- Most Economical Robust Selection Procedures for Location ParametersThe Annals of Statistics, 1979
- On two-stage selection procedures and related probability-inequalitiesCommunications in Statistics - Theory and Methods, 1978
- Note on Anderson's Sequential Procedures with Triangular BoundaryThe Annals of Statistics, 1974