Overview
Replete with easy-to-understand examples ranging from the prediction of home runs in baseball using an hierarchical Bayesian statistics model to estimating the expected return at blackjack using control variables, this text functions as a complete consideration of simulation. Sheldon Ross provides broad yet thorough coverage of the subject, presenting the development of a simulation study to analyze models, and demonstrates that by using random variables and the concept of discrete events, it is possible to generate the behavior of a stochastic model over time. Also discussed are questions concerning when to stop a simulation, how much confidence can be placed in the results, and extensive new information on the presentation of the alias method for generating discrete random variables material not found in any other text. Students, practitioners, and researchers alike will find this text to have an important place in their research libraries.Replete with easy-to-understand examples ranging from the prediction of home runs in baseball using an hierarchical Bayesian statistics model to estimating the expected return at blackjack using control variables, this text functions as a complete consideration of simulation. Sheldon Ross provides broad yet thorough coverage of the subject, presenting the development of a simulation study to analyze models, and demonstrates that by using random variables and the concept of discrete events, it is possible to generate the behavior of a stochastic model over time. Also discussed are questions concerning when to stop a simulation, how much confidence can be placed in the results, and extensive new information on the presentation of the alias method for generating discrete random variables material not found in any other text. Students, practitioners, and researchers alike will find this text to have an important place in their research libraries. Presents the statistics needed to analyze simulated data as well as those needed for validating the simulation model
Editorials
In the old days, says Ross (industrial engineering and operations research, U. of California-Berkeley), scientists formulating a stochastic model had to compromise between one that realistically portrayed the situation and one that could be comprehended mathematically. No more, he explains, now that fast and powerful computers are available cheap. He shows how to construct a model as faithful as possible to the phenomenon, then analyze it using a simulation study. He includes exercises for course study, but does not specify prerequisites. There's no mention of the dates of earlier editions. Annotation c. Book News, Inc., Portland, OR (booknews.com)