You are here

Foundations and Methods of Stochastic Simulation: A First Course

Barry L. Nelson
Publisher: 
Springer
Publication Date: 
2013
Number of Pages: 
276
Format: 
Hardcover
Series: 
International Series in Operations Research and Management Science 187
Price: 
89.95
ISBN: 
9781461461593
Category: 
Monograph
[Reviewed by
Robert W. Hayden
, on
03/20/2014
]

The most important thing to say about this book is that it originated as a textbook for a graduate simulation course in a Department of Industrial Engineering and Management Sciences. That affects the content in two important ways. The first is the choice of what to simulate: included here are queuing and network flow problems (broadly interpreted) but none of the sort of simulations done by statisticians or weather forecasters.

The second consequence of the book’s origin is in the choice of software. Simulations, by their nature, require software support. The two usual options are a dedicated simulation program and a general or scientific programing language. Regarding the first option, the author points out that “making it easy to model typical features can make it difficult to model something different, and research is always about something different.” So the author chooses to use Visual BASIC for Applications (VBA) as found in the commercial spreadsheet program Excel. The usual argument for using Excel in a business environment is that everyone already has it and knows how to use it. The usual argument against using Excel outside of accounting is that it is not a very good tool for other purposes. The author tries to address its limitations by providing a suite of VBA subs and class modules. These, as well as Java and MATLAB alternatives, are available on the book’s website. The book makes no effort to teach programing in any of these languages.

The intended audience is graduate and advanced undergraduate students. Realistic prerequisites might include prior programming experience in an object-oriented language, and a typical mathematical statistics course. The level of the text is a bit harder to pin down. The general approach is fairly abstract, with an emphasis on broad ideas applicable to a wide range of problems. Although few proofs or derivations are offered, many exercises ask students to produce them.

The book is clearly written, using a fairly terse style that may please instructors more than students. Perhaps its greatest strength is the attention it gives to the modeling process. For example, readers are encouraged to think long and hard about how to model inputs, rather than simply make the usual assumptions, or accept the defaults of their software. The author does this so well that this text can be recommended as a supplement to any simulation course. Whether to use it as a main text will depend largely on the degree of match between this book and the course syllabus and the intended audience. Where such a match exists, this text can be highly recommended for consideration.


After a few years in industry, Robert W. Hayden (bob@statland.org) taught mathematics at colleges and universities for 32 years and statistics for 20 years. In 2005 he retired from full-time classroom work. He now teaches statistics online at statistics.com and does summer workshops for high school teachers of Advanced Placement Statistics. He contributed the chapter on evaluating introductory statistics textbooks to the MAA's Teaching Statistics.