Empirical Bayes Methods for Discrete Event Simulation Performance Measure Estimation

Discrete event simulation (DES) is a widely-used operational research methodology facilitating the analysis of complex real-world systems. Although, generally speaking, simplicity is greatly desirable in DES modelling applications, in many cases the nature of the underlying system results in simulation models which are large in scale, complex, and expensive to run. As such, the careful design and analysis of simulation experiments is essential to ensure valid and efficient inference concerning DES model performance measures. It is envisaged that empirical Bayes (EB) methods, which enable data to be pooled across a set of populations to support inference of the parameters of a single population, may be of use within this context. Despite this potential, EB has so far been neglected within the DES literature. This paper presents a preliminary computational investigation into the efficacy of EB procedures in the estimation of DES performance measures. The results of this investigation, and their significance, are explored. Additionally, likely directions for future research are also addressed.

[1]  B. Efron,et al.  Limiting the Risk of Bayes and Empirical Bayes Estimators—Part II: The Empirical Bayes Case , 1972 .

[2]  Bradley P. Carlin,et al.  Bayesian Methods for Data Analysis , 2008 .

[3]  Isaac Dialsingh,et al.  Large-scale inference: empirical Bayes methods for estimation, testing, and prediction , 2012 .

[4]  D. Rasch.  Maritz, J. S.: Empirical BAYES Methods. Methuen's Monographs on Applied Probability and Statistics. Methuen and Co Ltd., London 1970. VIII f 159 s., 3 Abb., 32 Tab., Preis £2.50 , 1971 .

[5]  Christian P. Robert,et al.  Large-scale inference , 2010 .

[6]  Stewart Robinson,et al.  Conceptual modelling for simulation Part I: definition and requirements , 2008, J. Oper. Res. Soc..

[7]  B. Efron,et al.  Data Analysis Using Stein's Estimator and its Generalizations , 1975 .

[8]  Averill Law,et al.  Simulation Modeling and Analysis (McGraw-Hill Series in Industrial Engineering and Management) , 2006 .

[9]  H. Robbins.  An Empirical Bayes Approach to Statistics , 1956 .

[10]  Y. Benjamini.  Discovering the false discovery rate , 2010 .

[11]  Michael Pidd,et al.  Computer Simulation in Management Science (3rd Edition) , 1998 .

[12]  Jack P. C. Kleijnen,et al.  State-of-the-Art Review: A User's Guide to the Brave New World of Designing Simulation Experiments , 2005, INFORMS J. Comput..

[13]  Sheldon M. Ross,et al.  Introduction to probability models , 1975 .

[14]  Zhigen Zhao,et al.  Double shrinkage empirical Bayesian estimation for unknown and unequal variances , 2010 .

[15]  B. Efron,et al.  Stein's Estimation Rule and Its Competitors- An Empirical Bayes Approach , 1973 .