Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
|---|---|---|---|---|---|
Searching... | 30000004301200 | QA279 T564 2000 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
A unique, integrated treatment of computer modeling and simulation "The future of science belongs to those willing to make the shift to simulation-based modeling," predicts Rice Professor James Thompson, a leading modeler and computational statistician widely known for his original ideas and engaging style. He discusses methods, available to anyone with a fast desktop computer, for integrating simulation into the modeling process in order to create meaningful models of real phenomena. Drawing from a wealth of experience, he gives examples from trading markets, oncology, epidemiology, statistical process control, physics, public policy, combat, real-world optimization, Bayesian analyses, and population dynamics. Dr. Thompson believes that, so far from liberating us from the necessity of modeling, the fast computer enables us to engage in realistic models of processes in , for example, economics, which have not been possible earlier because simple stochastic models in the forward temporal direction generally become quite unmanageably complex when one is looking for such things as likelihoods. Thompson shows how simulation may be used to bypass the necessity of obtaining likelihood functions or moment-generating functions as a precursor to parameter estimation. Simulation: A Modeler's Approach is a provocative and practical guide for professionals in applied statistics as well as engineers, scientists, computer scientists, financial analysts, and anyone with an interest in the synergy between data, models, and the digital computer.
Author Notes
JAMES R. THOMPSON , PhD, is Professor of Statistics at Rice University. A Fellow of the American Statistical Association and the Institute of Mathematical Statistics, he is an elected member of the International Statistical Institute. In 1985, he received the ASA's Don Owen Award, and in 1991, he was awarded the U.S. Army's Samuel S. Wilks Medal for his work in applied statistics. A frequent consultant to industry, he holds adjunct professorships at the M. D. Anderson Cancer Center and the University of Texas School of Public Health. He is the author of ten books, including Empirical Model Building , available from Wiley.
Reviews 1
Choice Review
With the advent of faster computers with comparatively large storage facilities, simulation-based modeling is rapidly being adopted as an alternative approach to conventional top-down, assumptions-based, continuous, stochastic and discrete differential equation modeling. Thompson (Rice Univ.) offers an interesting exposition to the art of simulation, and views "simulation approach" to modeling as a paradigm for realistic evolutionary modeling. The book is written in a very casual style, and background knowledge in statistics is all that is required to grasp the material contained therein. Thompson begins with an exposition of the generation of random numbers and then delves into a variety of special topics, including models for stocks and derivatives, optimization and estimation in a noisy world, Monte-Carlo solutions to differential equations, simulation assessment of multivariate and robust procedures in statistical process control, resampling-based tests, and some exposition to modeling the AIDS epidemic. Several useful algorithms, problem sets, and references to standard simulation packages are provided. Short chapter bibliographies; wide range of examples. The book could be used as a resource for beginning graduate students and professionals in applied statistics, computer science, economics and finance, engineering, and the natural sciences. Highly recommended. Graduate students; faculty; professionals. D. E. Bentil; University of Vermont
Table of Contents
| The Generation of "Random" Numbers |
| Random Quadrature |
| Monte Carlo Solutions of Differential Equations |
| Markov Chains, Poisson Processes and Linear Equations |
| SIMEST, SIMDAT, and Pseudoreality |
| Models for Stocks and Derivatives |
| Simulation Assessment of Multivariate and Robust Procedures in Statistical Process Control |
| Noise and Chaos |
| Bayesian Approaches |
| Resampling Based Tests |
| Optimization and Estimation in a Noisy World |
| Modeling the USA AIDS Epidemic: Exploration, Simulation and Conjecture |
| Appendices |
| Index |
