Multivariate inferences for regenerative simulations to a specified precision level

ABSTRACT

When making an inference from a simulation output, the problem we usually encounter is the autocorrelation among observations. In this paper we proposed a technique using regenerative methods to overcome this problem and establish simultaneous confidence intervals for some correlated variables to a prespecified relative precision level. Five simulation models are used to test the performance of this technique and the coverage rate is the criterion. From the empirical results, the performance of this technique is quite satisfactory.

REFERENCES

• 1: Billingsley, P. (1979). Probability and Measure, Wiley, New York.
• 2: Carson, J. S. (1980). Linear Combinations of Indirect Estimators for Variance Reduction in Regenerative Simulation, Technical Report, School of Industrial and System Engineering, Georgia Institute of Technology.
• 3: Chen, Der-Fa Robert. (1985). Multivarate Inference in the Discrete Event Simulation Using Batch Means Method, Ph. D Dissertation, University of Georgia, Athens, Georgia.
• 4: Chen, Der-Fa Robert. (1986). Batch Means Method in the Steady-State Discrete Event Simulation, Proceeding of the 1986 CIIE national Conference, 165--190.
• 5: Cox, D. R. and Smith, W. L. (1961). Queues, Metheun London.
• 6: Crane, M. A. and Iglehart, D. L. (1974a). Simulating Stable Stochastic Systems I : General Multiserver Queues, J. Assoc. Comput. Mach. 21, 103--113.
• 7: Crane, M. A. and Iglehart, D. L. (1974b). Simulating Stable Stochastic Systems II : Markov Chains, J. Assoc. Comput. Mach. 21, 114--123.
• 8: Crane, M. A. and Iglehart, D. L. (1975a). Simulating Stable Stochastic Systems III : Regenerative Processes and Discrete-Event Simulations, Operations Research 23, 33--45.
• 9: Crane, M. A. and Iglehart, D. L. (1975b). Simulating Stable Stochastic Systems IV : Approximation Techniques, Management Science 21, 1215--1224.
• 10: Fishman, G. S. (1973a). Concepts and Methods in Discrete Event Digital Simulation, Wiley, New York.
• 11: Fishman, G. S. (1973b). Statistical Analysis for Queueing Simulations, Management Science 20, 363--369.
• 12: Fishman, G. S. (1974). Estimation in Multiserver Queueing Simulations, Operations Research 22, 72--78.
• 13: Fishman, G. S. (1978a). Grouping Observations in Digital Simulation, Management Science 24, 510--521.
• 14: Fishman, G. S. (1978b). Principles of Discrete Event Simulation, John Wiley and Sons, New York.
• 15: Law, A. M. and Kelton, W. D. (1982a). Confidence Intervals for Steady-State Simulation II : A Survey of Sequential Procedures, Management Science 28, 550--562.
• 16: Law, A. M. and Kelton, W. D. (1982b). Simulation Modeling and Analysis, McGraw-Hill, New York.
• 17: Liou, S. J. (1987). A Sequential Method in Multivariate Simulation Output Analysis, Master Thesis, Graduate School of Industrial Management, National Cheng Kung University, Tainan, Taiwan, R.O.C.
• 18: Morrison, D. F. (1976). Multivariate Statistical Method, John Wiley and Sons, New York.
• 19: Schruben, L. (1980). Control of Initialization Bias in Multivariate Simulation Response, Technique Report 455, School of Operations Research and Industrial Engineering, Cornell University.
• 20: Seila, A. F. (1982). Multivariate Estimation in Regenerative Simulation, Department of Quantitative Business Analysis, University of Georgia.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.6 SIMULATION AND MODELING

Additional Classification:
  G. Mathematics of Computing
  G.m MISCELLANEOUS
      Subjects: Queueing theory**

General Terms:
Algorithms, Design, Experimentation, Measurement, Performance