Advances in Stochastic Simulation Methods

This is a volume consisting of selected papers that were presented at the 3rd St. Petersburg Workshop on Simulation held at St. Petersburg, Russia, during June 28-July 3, 1998. The Workshop is a regular international event devoted to mathematical problems of simulation and applied statistics organized by the Department of Stochastic Simulation at St. Petersburg State University in cooperation with INFORMS College on Simulation (USA). Its main purpose is to exchange ideas between researchers from Russia and from the West as well as from other coun tries throughout the World. The 1st Workshop was held during May 24-28, 1994, and the 2nd workshop was held during June 18-21, 1996. The selected proceedings of the 2nd Workshop was published as a special issue of the Journal of Statistical Planning and Inference. Russian mathematical tradition has been formed by such genius as Tchebysh eff, Markov and Kolmogorov whose ideas have formed the basis for contempo rary probabilistic models. However, for many decades now, Russian scholars have been isolated from their colleagues in the West and as a result their mathe matical contributions have not been widely known. One of the primary reasons for these workshops is to bring the contributions of Russian scholars into lime light and we sincerely hope that this volume helps in this specific purpose.

Auteur
N. BALAKRISHNAN, PhD, is Professor of Mathematics and Statistics at McMaster University in Hamilton, Ontario, Canada.V. B. NEVZOROV, PhD, DS, is Professor of Probability and Statistics at St. Petersburg State University in St. Petersburg, Russia.

Texte du rabat
This carefully edited book discusses new methods and applications for stochastic simulation and experimental design with the focus on methodological issues and recent developments for computer simulations in statistical application problems. A large number of topics are treated, including computer simulation methodology queueing systems, statistical methods in simulation, optimal experimental design, and numerical algorithms. The book will be an essential up-to-date resource for researchers and professionals in applied statistics, experimental design, operations research and stochastic simulation.

Contenu
I: Simulation Models.- 1 Solving the Nonlinear Algebraic Equations with Monte Carlo Method.- 2 Monte Carlo Algorithms For Neumann Boundary Value Problem Using Fredholm Representation.- 3 Estimation Errors for Functionals on Measure Spaces.- 4 The Multilevel Method of Dependent Tests.- 5 Algebraic Modelling and Performance Evaluation of Acyclic Fork-Join Queueing Networks.- II: Experimental Designs.- 6 Analytical Theory of E-Optimal Designs for Polynomial Regression.- 7 Bias Constrained Minimax Robust Designs for Misspecified Regression Models.- 8 A Comparative Study of MV- and SMV-Optimal Designs for Binary Response Models.- 9 On the Criteria for Experimental Design in Nonlinear Error-In-Variables Models.- 10 On Generating and Classifying all q71-m-1Regularly Blocked Factional Designs.- 11 Locally Optimal Designs in Non-Linear Regression: A Case Study of the Michaelis-Menten Function.- 12 D-Optimal Designs for Quadratic Regression Models.- 13 On the Use of Symmetry in Optimal Design of Experiments.- III: Statistical Inference.- 14 Higher Order Moments of Order Statistics from the Pareto Distribution and Edgeworth Approximate Inference.- 15 Higher Order Moments of Order Statistics from the Power Function Distribution and Edgeworth Approximate Inference.- 16 Selecting from Normal Populations the One with the Largest Absolute Mean: Comon Unknown Variance Case.- 17 Conditional Inference for the Parameters of Pareto Distributions when Observed Samples are Progressively Censored.- IV: Applied Statistics and Related Topics.- 18 On Randomizing Estimators in Linear Regression Models.- 19 Nonstationary Generalized Automata with Periodically Variable Parameters and Their Optimization.- 20 Power of Some Asymptotic Tests for Maximum Entropy.- 21 Partially Inversion of Functions forStatistical Modelling of Regulatory Systems.- 22 Simple Efficient Estimation for Three-Parameter Lognormal Distributions with pplications to Emissions Data and State Traffic Rate Data.