Abstract
This paper provides an overview of the five most commonly used statistical techniques for improving the efficiency of stochastic simulations: control variates, common random numbers, importance sampling, conditional Monte Carlo, and stratification. The paper also describes a mathematical framework for discussion of efficiency issues that quantifies the trade-off between lower variance and higher computational time per observation.
Similar content being viewed by others
References
S. Asmussen, Busy period analysis, rare events, and transient behaviour in fluid models, J. Appl. Math. Stoch. Anal. (1993), to appear.
P. Bratley, B.L. Fox and L.E. Schrage,A Guide to Simulation (Springer, New York, 1987).
J. Bucklew,Large Deviation Techniques in Decision, Simulation, and Estimation (Wiley, New York, 1990).
C.S. Chang, P. Heidelberger, S. Juneja and P. Shahabuddin, Effective bandwidth and fast simulation of ATM intree networks, IBM Research Report RC 18586, Yorktown Heights, New York (1992), in:Proc. Performance 1993 Conf., to appear.
R.C.H. Cheng, The use of antithetic variates in computer simulations, J. Oper. Res. Soc. 21(1982)229–237.
W.G. Cochran,Sampling Techniques (Wiley, New York, 1977).
G.S. Fishman and B.D. Huang, Antithetic variates revisited, Commun. ACM 26(1983)964–971.
G.S. Fishman and V.G. Kulkarni, Improving Monte Carlo efficiency by increasing variance, Manag. Sci. 38(1992)1432–1444.
B.L. Fox and P.W. Glynn, Discrete-time conversion for simulating semi-Markov processes, Oper. Res. Lett. 4(1986)49–53.
B.L. Fox and P.W. Glynn, Discrete-time conversion for simulating finite-horizon Markov processes, SIAM J. Appl. Math. 50(1990)1457–1473.
P. Glasserman, Processes with associated increments, J. Appl. Prob. 29(1992)313–333.
P. Glasserman, Stochastic monotonicity and conditional Monte Carlo for likelihood ratios, Adv. Appl. Prob. 25(1993)103–115.
P. Glasserman, Filtered Monte Carlo, Math. Oper. Res. 18(1993)610–634.
P. Glasserman and S.-G. Kao, Overflow probabilities in Jackson networks,Proc. 32nd IEEE Conf. on Decision and Control (1993), pp. 3178–3182.
P. Glasserman and D. Yao, Some guidelines and guarantees for common random numbers, Manag. Sci. 38(1992)884–908.
P.W. Glynn, Regenerative structure of Markov chains simulated via common random numbers, Oper. Res. Lett. 4(1985)49–53.
P.W. Glynn and D.L. Iglehart, Simulation methods for queues: An overview, Queueing Syst. 3(1988)221–256.
P.W. Glynn, Optimization of stochastic systems via simulation,Proc. 1989 Winter Simulation Conf. (1989) pp. 90–105.
P.W. Glynn, A GSMP formalism for discrete-event systems, Proc. IEEE 77(1989)14–23.
P.W. Glynn and D.L. Iglehart, Importance sampling for stochastic simulation, Manag. Sci. 35(1989)1297–1325.
P.W. Glynn and W. Whitt, Indirect estimation viaL = λW, Oper. Res. 37(1989)82–103.
P.W. Glynn and W. Whitt, Asymptotic efficiency of simulation estimators, Oper. Res. 40(1992)505–520.
A. Goyal, P. Shahabuddin, P. Heidelberger, V.F. Nicola and P.W. Glynn, A unified framework for simulating Markovian models of highly reliable systems, IEEE Trans. Computers C-41(1992)36–51.
J.M. Hammersley and D.C. Handscomb,Monte Carlo Methods (Methuen, London, 1964).
P. Heidelberger, Fast simulation of rare events in queueing and reliability models, IBM Research Report RC 19028, Yorktown Heights, NY (1993).
P. Heidelberger and D.L. Iglehart, Comparing stochastic systems using regenerative simulations with common random numbers, Adv. Appl. Prob. 11(1979)804–819.
S.S. Lavenberg, T.L. Moeller and P.D. Welch, Statistical results on multiple control variables with application to queueing network simulation, Oper. Res. 30(1982)182–202.
A.M. Law, Efficient estimators for simulated queueing systems, Manag. Sci. 22(1975)30–41.
B. Nelson, An illustration of the sample space definition of simulation and variance reduction, Trans. Soc. Comput. Simul. 2(1985)237–247.
B. Nelson, Decomposition of some well-known variance reduction techniques, J. Statist. Comput. Simul. 23(1986)183–209.
D. Ruppert, Almost sure approximations to the Robbins-Monro and Kiefer-Wolfowitz procedure with dependent noise, Ann. Prob. 10(1982)178–187.
K.W. Ross and J. Wang, Monte Carlo summation applied to multichain queueing networks,Proc. 30th IEEE Conf. on Decision and Control (1991) pp. 883–884.
J.S. Sadowsky, Large deviations theory and efficient simulation of excessive backlogs in aGI/G/m queue, IEEE Trans. Auto. Contr. AC-36(1991)1383–1394.
P. Shahabuddin, Importance sampling for the simulation of highly reliable Markovian systems, IBM Research Report RC 16729, Yorktown Heights, New York (1993), Manag. Sci., to appear.
J.R. Wilson, Variance reduction techniques for digital simulation, Amer. J. Math. Manag. Sci. 4(1984)277–312.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Glynn, P.W. Efficiency improvement techniques. Ann Oper Res 53, 175–197 (1994). https://doi.org/10.1007/BF02136829
Issue Date:
DOI: https://doi.org/10.1007/BF02136829