Reviewed by:
Timothy E. McMahon, M.S.
Electronic Publishing Specialist
American Mathematical Society

A researcher approaches the reference desk and asks how she can find the latest figures on terrorist incidents in the United States. Another approaches asking to learn how much money was spent to make the movie "Independence Day." Can you answer these questions using online resources that are available from your library? Possibly. However, if you have read Berinstein's book, that answer quickly becomes Probably.

Paula Berinstein has put together a comprehensive work that covers the Internet as well as other fee based online services. The author leads off her work with a detailed Table of Contents and Table of Figures. These are complimented by a pointer to the Directory of Online Statistical Sources locate on the Berinstein Research Web Site. This site is a companion to the book and attempts to keep users up-to-date on trends effecting the discovery and use of online statistical sources. The book contains nineteen chapters, four appendices and a solid forty-three page index. Throughout the book, the author uses a conversational tone with her readers that serves to ease the user into the complexities of statistics discovery and retrieval.

The first chapter of the book is a Quick Start designed for those familiar with online searching and statistics. This chapter contains brief abstracts as well as general and specific online services that are commonly available. The reader is presented sections dealing with availability, costs, features and benefits as well as drawbacks of these sites. The chapter also includes several comparison charts and tips on search strategy.

Chapter two is a primer on statistics. Here the reader will receive a brief overview of the types of statistics one might find online. Concepts such as raw numbers, percentages, averages and standard deviation are clearly defined on a low level so as to make these abstractions available to the broadest audience. The author also discusses methodologies of data collection and analysis that are common in statistical reporting.

Chapter three concentrates on giving the reader an understanding of who generates and publishes statistics. These providers are broken out into fifteen broad categories that range from government agencies to individual researchers. The author takes a brief look at each of these categories with much of the text devoted to the federal government. This chapter provides easy to read bulleted lists that point to sites where the user will find statistical information and screen captures that allow the reader to see what he or she should be looking at when using the links provided by the author.

The fourth chapter provides the reader with general search tips. In this chapter, the author presents the reader with tips for choosing statistical sources and combines these with advice on formulating search strategies. The core of this chapter is the author's construction of figures and word lists that cause the reader to think not only of where to find statistics but how statistics might be presented on any site or service. Also provided in this chapter are tips on searching specific services such as DIALOG, STN, DataStar and others.

Chapters five through eighteen are "subject-specific" and deal with finding statistics in particular subject areas. These subject areas range from demographics and population to transportation statistics. Each of these chapters details common types of data to be found, key producers of this data, best places to find these data and an extremely useful case study. These case studies present the reader with a reference question and methodically steps through the process involved in the discovery of a correct answer. Users will see the purpose of the exercise, reference question, likely sources to use, access points used to find the data and the system where the searcher discovered the answer. The author then reviews the search methodology and presents figures captured from the search service to augment the discussion.

Following her summary, Berinstein presents the reader with four appendices designed to round-out the content of the book. Appendix A provides contact information for a set of information providers, many of which were mentioned throughout the book. Appendix B lists the case studies presented in this book while Appendix C presents a useful glossary of statistical terms. The final appendix contains a bibliography of works useful to the author in construct of the book. This is also to be seen as a "further reading" section. As mentioned earlier, the index is comprehensive and easy to use.

In this information age we are all experiencing information overload in two aspects: 1. Searching for information takes longer using the new internet technologies vs. the traditional online systems, and 2. We get too much recall with too little precision. A handy tool to overcome these problems is Paula Berinstein's book "Finding Statistics Online". This well-organized reference is a true problem-solver for information seekers who value their time and need to find statistical data for marketing or research purposes. Finding Statistics Online starts with three very important chapters that explain how data is gathered, the various sources of statistical data and a comparison of online vendors including excellent search tips. All other 16 chapters of the book are arranged by topic such as business, health, education etc. and consist of a description of the typical data, key producers and resources, best places to look and a case study. This book gives the information searcher various options for each subject matter including free internet resources as well as commercial vendors. Finding Statistics Online is a highly recommended reference for every searcher.

Markov chains, a sequence of actions with a random element, can be applied to nearly any process that is composed of discrete steps. Such processes include manufacturing operations, building large structures and even visiting web sites on the Internet. This book is a set of lecture notes for a course for advanced undergraduates offered by the author.
It begins with a basic introduction to probability theory and covers some applications of Markov chains that I had never thought about, such as surfing the Internet. There are exercises at the end of each section although solutions are not included. With a section on computer simulation of Markov chains, the book could also be used in a special topics course in computer science, where the emphasis is on optimization.
Markov chains are a valuable tool that should receive more emphasis in the education of math and computer science students. This book can serve as a basic text for classes aimed at both types of student.

This is a very clearly written, succinct introduction to Markov Chains (it does not aim to be exhaustive).
I think this book offers one of the most accessible and efficient routes to learning the basics about markov chain monte carlo, perfect simulation,sandwiching and simulated annealing.

I enjoyed this book a lot.Better to read this a few times and do the questions, and then if you must, tackle one of the more bloated expositions.

This review book is formated in such a way that I can actually help my child with homework. I go directly to the section she's learning, and can quickly review the section that's current to her learning needs. She and I can do the reviews together and this wouldn't be possible without this book. I especaily like that I don't have to refer to hundreds of pages before being able to help my child.

This is an excellent refresher for Statistics, laid out logically and with ideas clearly illustrated graphically. Published by Barron's, these "Forgotten" books are nicely oriented toward the professional who needs to brush up on a subject in order to make competent decisions and communicate intelligently.

"Statistics and Probability" is one of those amazing sciences that we take for granted from childhood, but the more we think about it (as we get older, regrettably), it begins to dawn on us that this is the catalyst of many mysterious processes, such as the forces of life overcoming entropy. Science has rather recently discovered that there is really no such thing as "randomness", with the advent of Chaos Theory bringing down the walls of the last bastion of that idea. Statistics and Probability then become the focus of means by which some form of Higher Intelligence manages to do its business under our noses without us having a clue as to its presence!

The author largely explores information from the vantagepoint of its relationship with statistics. Early on, the basic properties of information are discussed. Information is then illustrated and contextualized with respect to specific probability distributions.

This is a very well written text by Kullback. Kullback is well known for the famous Kullback-Liebler information measure of distance between distribution functions. This was the first statistics text to cover statistical methods from the information theory viewpoint and now stands as a classic. For more recent developments see the text by Cover and Thomas. Rissanen also has an interesting book that shows how he used information theory ideas in model selection problems. Information theory goes back to Claude Shannon and others who worked on problems of encoding information for efficient transmission (particularly early telephone applications). This has been very important in electronic communications and is growing in use with modern satellite transmissions and the growing use of computer networks.

This is an excellent book covering the latest thinking in credit risk modeling. It could have used more on documentation and structuring risk, so I only gave it four stars, but it is five stars in model review.

For more on product descriptions and structuring risk, I highly recommend Tavakoli's "Credit Derivatives" 2nd Edition.

This is an excellent treatise on the near state-of-the-art in credit risk management. Although the focus is on sell-side risk management, many (if not all) of the techniques described can be used on the buy-side also.

This is the first book that really focusses on the portfolio problem of credit risk - many books have touched on vendor-provided models and their shortcomings but Bluhm et al. take it further into the practitioner's world.

The reader does not need a very strong background in math or physics but some understanding of finance and stochastic calculus would help to get the most out of it.

I recommend to everyone who is either in or thinking of getting into credit risk as a career - enjoy....

The first edition of Ted Anderson's text on multivariate analysis was published in 1959. At the time it had no rivals. This book gives a thorough mathematical treatment of classical multivariate analysis. It is extremely well organized. Development of the multivariate normal distribution and its properties are given a thorough and rigorous treatment. The Wishart distribution is derived. Properties of the multivariate normal distribution are applied to problems of classification, principal components, canonical correlation and tests of hypotheses including the use of Hotelling's T square.

As a graduate student at Stanford, I audited Ted Anderson's multivariate analysis course, that he taught out of the first edition of the book. It wasn't until 1984 that he revised the text incorporating some new materials including the bootstrap method.

This is an advanced course for graduate students in statistics. It is the best source for a rigorous mathematical treatment of the important results from the theory of the multivariate normal distribution. However, it is not easy reading for someone who is interested in applications but does not have strong training in mathematics (particularly linear algebra). For applications and approaches when the normal theory doesn't apply, the book by Gnanadesikan is very good. There are now many good theoretical and applied texts on multivariate analysis including the text by Eaton, the one by Srivastava and Khatri, one by Rencher, one by Johnson and Wichern, and the one by Mardia, Kent and Bibby. Naik and Khattree have written a very nice applied multivariate book that demonstrates the applications using SAS software every step of the way.

There are now many subspecialties including cluster analysis, principal components, correspondence analysis, factor analysis and classification that have complete texts devoted to them.

A non-esoteric introduction into the discipline of multivariate statistical theory. Accessible with undergraduate-level mathematics while retaining much of the important "guts". It's a shame that the Wiley series is often tres expensive, as opposed to budget books like Dover.

The best thing about this book is that it teachs basic probability theory while keeping the reader constantly aware of the on-going debate regarding what it means to talk in terms of probabilities, and of how that debate has shaped the development of probability theory. If you are a student taking a course in probability and statistics who would like to genuinely understand the conceptual basis of all those formulas they are teaching you, I suggest you read this book.

Some readers will be disappointed by this book. Since the book concentrates on the conceptual basis of probability and inductive logic, it does not give the reader enough technical tools to really do much applied mathematics. On the other hand, by the time Hacking gets around to discussing what students of philosophy will likely view as the big philosophical pay-off of probability theory (i.e. Bayesian and frequentist contributions to the problem of justifying induction) he devotes to them a mere 20 pages of not terribly deep discussion.

I would HIGHLY recommend this book for anyone (including business men) who must make decisions with incomplete information and under uncertainty. Instead of focusing on the mechanics of statistics, it focuses on how to think about risky propositions.

I bought this book while working on a particular problem in machine learning, at a point where I had started realizing that I was losing clarity on my definition of probability. I was using the mechanics, but didn't clearly understand why the use was valid. This seemed an odd and embarrassing circumstance at the time, how could I not understand what "probability" means? As it turns out this confusion is one shared broadly in history of science, and in current applications of statistical mechanics.

Prof Hacking's writing is clear and entertaining, clearly aimed at engaging the reading audience.

This book is the most current text available on the theory of extreme values. The author eloquently provides us with an understanding of the theory and it vast applications. It is intended for researchers students and practitioners. So it provides an in-depth account of the theory with many real world examples. It contains an excellent up-to-date bibliography. Important theorems are presented with their implications but without mathematical proofs. Computations are done in SPlus. The author provides an appendix on computational aspects that tells the reader where to go to download examples and find the SPlus functions that are used.

Topics include classical extreme value theory and models, threshold models, extremes in dependent stationary cases, extremes for some nonstationary stochastic processes, the point process approach, multivariate extremes and some special topics including extremes in spatial processes and the Bayesian approach to extremes (with examples employing MCMC methods).

I recently used the software accompanied to this book kindly made available by the author and was led to know more about this book and the author's other works. I like what I saw and think the author has done a supeb job in explaining the difficult theory in plain language and in the context of data analysis. Thus it is an "action" book instead of the "just theory" as with most other books. The book provides a balanced treatment of different approaches to extreme value analysis. Personally I prefer the generalized Pareto approach, though theoretically the point process approach may be very neat, if it can be realized.
I think extreme value theory in general is an important statistical area, since in practice one may be forced to deal with analyzing extreme events, such as in financial engineering, environmental or climate analysis, or network design. I wholeheartedly recommend this book for anyone who want to learn this area from one of the leading researchers.

Fukunaga is a standard source for pattern recognition methods often cited in the engineering literature. Covers parametric (particularly linear and quadratic discriminant algorithms) and nonparametric methods (density estimation). It is designed for and popular with engineers. When I was working at Nichols Research Corporation Fukunaga's papers and this book (earlier edition) were often cited as sources to justify the algorithms we used for discrimination problems. In fact Fukunaga had been a consultant to the company (used primarily by the Boston branch of the company where the KENN algorithms were developed). It is a reputable source. I still like Duda and Hart (1972) for good explanations of the fundamental concepts. For statisticians McLachlan's book is now far and away the best source.

If you are writing a machine learning paper, and need to cite something to support an argument, you can almost always cite Fukunaga. This work is a standard reference in the field. The presentation of most material is very terse, but that is great if you already have a good feel for the material and need to look up some details about some algorithm or technique. There isn't much about neural networks here, but for the rest of the pattern recognition techniques, this is almost always the first place to start. Another strong point for this book is the use of realistic examples, which illustrate many of the statistical techniques.

To prepare for my first semester teaching statistics, I asked several colleagues for advice and references. Texts by Brunk and DeGroot were very useful and helped me organize and clarify for my students the material in our "statistics for engineers" text. Recently I discovered the three wonderful volumes by Hoel, Port, and Stone. In particular I found nice pictures to illustrate and explain the central limit theorem. There are also many exercises and worked examples. The next statistics course I teach will be greatly improved thanks to these books and the work that went into writing them.

When I was a graduate student at Stanford in the mid 1970s I took my first course in stochastic processes out of the companion book on stochastic processes by these authors. The three volumes "Introduction to Probability", "Introduction to Statistics" and "Introduction to Stochastic Processes" were produced at about the same time with the intention of covering the first three graduate courses on these subjects in a statistics porgram. I had different texts for my probabilty and statistics courses but these three texts were excellent and very similar in style. They are all clear and concise and could be best studied in the order given above.