A Sequential Procedure for Determining the Length of a Steady-State Simulation

A common problem faced by simulators is that of constructing a confidence interval for the steady-state mean of a stochastic process. We have reviewed the existing procedures for this problem and found that all but one either produce confidence intervals with coverages which may be considerably lower than desired or have not been adequately tested. Thus, in many cases simulators will have more confidence in their results than is justified. In this paper we present a new sequential procedure based on the method of batch means for constructing a confidence interval with coverage close to the desired level. The procedure has the advantage that it does not explicitly require a stochastic process to have regeneration points. Empirical results for a large number of stochastic systems indicate that the new procedure performs quite well.

[1]  Carl M. Harris,et al.  Fundamentals of queueing theory , 1975 .

[2]  D. Daley.  The serial correlation coefficients of waiting times in a stationary single server queue , 1968, Journal of the Australian Mathematical Society.

[3]  George S. Fishman,et al.  Solution of Large Networks by Matrix Methods , 1976, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  Benjamin Avi-Itzhak,et al.  A Time-Sharing Queue with a Finite Number of Customers , 1969, JACM.

[5]  Rupert G. Miller.  The jackknife-a review , 1974 .

[6]  T. Anderson.  Statistical analysis of time series , 1974 .

[7]  Michael A. Crane,et al.  Simulating Stable Stochastic Systems, I: General Multiserver Queues , 1974, JACM.

[8]  D. Iglehart.  Simulating stable stochastic systems, V: Comparison of ratio estimators , 1975 .

[9]  Robert G. Sargent,et al.  Statistical analysis of simulation output data , 1976, SIML.

[10]  Michael A. Crane,et al.  Simulating Stable Stochastic Systems, II: Markov Chains , 1974, JACM.

[11]  Averill M. Law.  Confidence intervals in discrete event simulation: A comparison of replication and batch means , 1977 .

[12]  R. R. Coveyou,et al.  Fourier Analysis of Uniform Random Number Generators , 1967, JACM.

[13]  Averill M. Law.  Efficient estimators for simulated queueing systems , 1974 .

[14]  George S. Fishman,et al.  Statistical Analysis for Queueing Simulations , 1973 .

[15]  Jeffrey P. Buzen,et al.  Queueing Network Models of Multiprogramming , 1971, Outstanding Dissertations in the Computer Sciences.

[16]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1972 .

[17]  George S. Fishman,et al.  Achieving specific accuracy in simulation output analysis , 1977, CACM.

[18]  C. H. Sauer,et al.  Sequential stopping rules for the regenerative method of simulation , 1977 .