Sensitivity Analysis for Simulations via Likelihood Ratios
Operations ResearchPublished 1 October 1989
Martin I. Reiman, Alan Weiss
Citations264
SJR quartileQ1
SJR score2.56
SNIP1.83
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A simple method of estimating the sensitivity of quantities obtained from simulation with respect to a class of parameters, which includes Poisson rates, discrete probabilities, and the mean and variance of a Normal distribution is presented.
Abstract
We present a simple method of estimating the sensitivity of quantities obtained from simulation with respect to a class of parameters. Here sensitivity means the derivative of an expectation with respect to a parameter. The class of parameters includes, for example, Poisson rates, discrete probabilities, and the mean and variance of a Normal distribution. The method is extremely well suited to regenerative simulation, and can be implemented on extant simulations with little effort, increase in running time, or memory requirements. It is based on some change-of-measure formulas derived from likelihood ratios. In addition to the theorems that underly the technique, we present some numerical examples.
Keywords
Decision SciencesMathematics
The Annals of Mathematical StatisticsA Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on the sum of Observations
3,567 Citations1952Herman Chernoff
Monte Carlo Methods
3,037 Citations1964J. M. Hammersley, D. C. Handscomb
Journal of the American Statistical AssociationMathematical Statistics: Basic Ideas and Selected Topics.
1,915 Citations1979Kenneth W. Wachter, Peter J. Bickel +1 more
The Annals of StatisticsImportance Sampling in the Monte Carlo Study of Sequential Tests
332 Citations1976David Siegmund
Management SciencePerturbation Analysis Gives Strongly Consistent Sensitivity Estimates for the <i>M</i>/<i>G</i>/1 Queue
238 Citations1988Rajan Suri, Michael A. Zazanis
IEEE Transactions on Automatic ControlSmoothed (conditional) perturbation analysis of discrete event dynamical systems
226 Citations1987Weibo Gong, Yu‐Chi Ho
A new sample path analysis approach based on the smoothing property of conditional expectation for estimating the performance sensitivity of discrete event dynamical systems is proposed and it is pointed out that one can get more knowledge about the system performance by observing and analyzing the sample path than by using the conventional simulation approach.
IEEE Transactions on Automatic ControlLarge deviations and rare events in the study of stochastic algorithms
211 Citations1983M. Patrick Cottrell, Jean‐Claude Fort +1 more
New asymptotics formulas for the mean exit time from an almost stable domain of a discrete-time Markov process are obtained and an original fast simulation method is proposed based on the classical Robbins-Monroe algorithm.
Operations ResearchSensitivity Analysis and Performance Extrapolation for Computer Simulation Models
183 Citations1989Reuven Y. Rubinstein
It is shown that both the sensitivities (derivatives, gradients, Hessians, etc.) and the performance measure can be estimated simultaneously from the same simulation.
Management ScienceConvergence Properties of Infinitesimal Perturbation Analysis Estimates
164 Citations1988Philip Heidelberger, Xi‐Ren Cao +2 more
Mathematics of Operations ResearchOpen Queueing Systems in Light Traffic
119 Citations1989Martin I. Reiman, Burton Simon
Focusing initially on Poisson arrival processes, a method is provided to calculate fn0 for any 'admissible' function f, and a large class of queueing networks for which several standard quantities of interest are shown to be admissible.
Winter Simulation ConferenceStochastic approximation for Monte Carlo optimization (1986)
104 Citations1986Peter W. Glynn
Bell System Technical JournalA Class of Closed Markovian Queuing Networks: Integral Representations, Asymptotic Expansions, and Generalizations*
76 Citations1981James McKenna, Debasis Mitra +1 more
Operations ResearchAn Asymptotic Analysis of a Queueing System with Markov-Modulated Arrivals
74 Citations1986David Y. Burman, Donald R. Smith
It is shown that in light and heavy traffic these quotients converge to finite, nonzero quantities that can be calculated in terms of the original model parameters.
Journal of the ACMRegenerative Simulation of Response Times in Networks of Queues
71 Citations1978Donald L. Iglehart, Gerald S. Shedler
The marked job simulation via hitting times and the decomposition method are introduced, which simplifies the simulation of regenerative processes and improves the efficiency of simulation.
Operations ResearchAn Interpolation Approximation for Queueing Systems with Poisson Input
64 Citations1988Martin I. Reiman, Burton Simon
The details of the above approximation are presented, focusing, via several examples, on applications of the method, and the results are compared to some related approximations.
Journal of Statistical Computation and SimulationSensitivity estimates based on one realization of a stochastic system†
38 Citations1987Xi‐Ren Cao
IEEE Transactions on Information TheoryLight traffic derivatives via likelihood ratios
25 Citations1989Martin I. Reiman, Alan Weiss
The steady-state behavior of open queuing systems with Poisson arrival processes in light traffic, that is, as the arrival rate tends to zero, is considered and the derivatives can be evaluated analytically in simple cases and by simulation in general.
AT&T Bell Laboratories Technical JournalAn Approximate Analysis of Sojourn Times in the M/G/1 Queue With Round-Robin Service Discipline
9 Citations1984Philip J. Fleming
This paper investigates the effect of the quantum size on the mean and variance of the response time in time-shared computer systems.
Simterpolations
6 Citations1987Martin I. Reiman, Burton Simon +1 more
An overview of a method for estimating an entire queueing function, ƒ(λ), 0 ≤ λ < c, from a single simulation sample path, where λ is the arrival rate to the system and c is the system capacity.