Performance evaluation for the score function method in sensitivity analysis and stochastic optimization

  • Søren Asmussen
  • Reuven Rubinstein
Conference paper
Part of the Lecture Notes in Economics and Mathematical Systems book series (LNE, volume 374)


Estimating systems performance 𝓁(ρ) = 𝔼 ρ L and the associated sensitivity (the gradient ∇𝓁(ρ)) for several scenarios via simulation generally requires a separate simulation for each scenario. The score function (SF) method handles this problem by using a single simulation run, but little is known about how the estimators perform. Here we discuss the efficiency of the SF estimators in the setting of simple queueing models. In particular we consider heavy traffic (diffusion) approximations for the sensitivity and the variances of the associated simulation estimators, and discuss how to choose a ‘good’ reference system (if any) in order to obtain reasonably good SF estimators.


Score Function Traffic Intensity Importance Sampling Variance Reduction Heavy Traffic 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    S. Asmussen (1987) Applied Probability and Queues, John Wiley & Sons.Google Scholar
  2. [2]
    S. Asmussen (1991) Queueing simulation in heavy traffic. Math. Opns. Res. Google Scholar
  3. [3]
    S. Asmussen & R.Y. Rubinstein (1992) The efficiency and heavy traffic properties of the score function method for sensitivity analysis of queueing models. Adv. Appl. Probab. Google Scholar
  4. [4]
    S. Asmussen & R.Y. Rubinstein (1991) The performance of likelihood ratio estimators using the score function. Stock. Models, (under revision).Google Scholar
  5. [5]
    P. Glasserman (1990) Stochastic monotonicity, total positivity, and conditional Monte Carlo for likelihood ratios. Manuscript, AT & T Bell Laboratories, Holmdel, N.J.Google Scholar
  6. [6]
    R.Y. Rubinstein (1989) Sensitivity analysis of computer simulation models via the efficient score. Opns.Res. 37, 72–81.CrossRefGoogle Scholar
  7. [7]
    R.Y. Rubinstein (1989b) Monte Carlo Methods for performance evaluation, sensitivity analysis and optimization of stochastic systems, Encyclopedie of Computer Science and Technology (Kent ed.) (to appear).Google Scholar
  8. [8]
    R.Y. Rubinstein & A. Shapiro (1991) Discrete Event Systems: Sensitivity Analysis and Stochastic Optimization via the Score Function Method, to be published by John Wiley & Sons.Google Scholar
  9. [9]
    W. Whitt (1989) Planning queueing simulations. Managmt. Sci. 35, 1341–1366.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Søren Asmussen
    • 1
  • Reuven Rubinstein
    • 2
  1. 1.Chalmers University of TechnologyGöteborgSweden
  2. 2.TechnionHaifaIsrael

Personalised recommendations