I bought this book as a "do it yourself" introduction to mathematics (for a biologist). It is quite rare to find a textbook as easy and enjoyable to follow as Frederick Adler's book. The text starts with very little assumed knowledge of mathematics and works through the basics of calculus using interesting examples taken from biology. The exercises at the end of each chapter are straightforward and easy to follow (my only complaint about the book is that they only provide solutions to every second question, so working alone it is hard to check your results). It is rare to find a text (especially in maths) that you can truly work through on your own. I recommend it for any Biologist wishing an intorduction to or to expand the knowledge of the growing field of Mathematical Biology.

I have enjoyed the "Introductory Statistics A Modelling Approach" by J. K. Lindsey very much, then it has been inducing me going to search more statistical books of J. K. Lindsey. Of course, I could not miss out this "Modelling Frequency and Count Data", as an actuarial student, I am so beneficial from reading and studying this text. It presents the standard models and many newly developed ones in a language that can be immediately apllied in many modern statistical computing package.

I recommend J. K. Lindsey's books to everyone who needs statistical knowledge and technique. "Modelling Frequency and Count Data" is an excellent book that you can't miss out!

When I looked at the cover of this book, I knew it was going to be contrarian. It shows volatility moving in the opposite direction to growth. Looking inside the book, I found the cover figure in Chapter 9 where it was described as volatility versus growth for 75 years of the Ibbotson Index starting in 1926. I looked at the Ibbotson table, and, indeed with the authors' eleven outlier years eliminated, the correlation is still negative (-.142). (With all 75 years included, I found the correlation to be (-.317).) At any rate, the Markowitzian notion of finding how large you can stand for volatility to be and then finding the portolio which maximizes growth is stood on its head.

The authors come up with an alternative to the Markowitz approach for portfolio selection based on something they call a simugram, which looks to be computer intensive.

Much of the book is spent on fundamental analysis, and indeed the authors do not seem favorably disposed to technical analysis. They dump on Black-Scholes and blame its use for the collapse of LTCM and Enron.

Some finance professionals will find much of this book annoying, since it attacks many standard concepts, such as the Efficient Market Hypothesis. And it seems to attack some of the basic tools in the finance tool kit, such as "risk neutral" evaluation.

One of the troubling things I found is that though the authors attack the canon of modern finance, they have only limited alternatives to recommend. They seem to recommend either doing deep fundamental analysis, using their complex simugram portfolio analysis, or putting one's money into an index fund. Most of us don't have the time to do the first or the software to do the second. To do the third really gives up on mathematical finance.

Usually books on regression analysis deal with only linear regression. This text includes thorough coverage of the standard regression topics but also includes chapters on logistic regression, nonparametric regression, robust regression, ridge regression and nonlinear regression. Recent tools including the bootstrap are covered. Software for ridge regression is discussed and a Minitab macro is given in Appendix A. Software is also provided in the attached diskette with installation instructions in Appendix A.

Ryan is an excellent writer who does not assume a high level of mathematical sophistication. He has also written an earlier text on statistical quality control "Statistical Methods for Quality Improvement" which came out in a second edition earlier this year.

This book covers a wide range of applications in Statistical Mechanics, with clear explanation, examples, tips, algorithms, and explicit programs at the end of the book. It is good for beginners and experienced alike, since it discusses "classical" and modern algorithms. It is a must for those who want to make actual numerical calculations in Statistical Physics.

The author is a top young gun from Harvard's Statistics Dept., and is an expert in many applied areas that utilize Monte Carlo, like the red hot bioinformatics. This book covers MC techniques developed in many different fields e.g., physics,structural biology, statistics. It has a wide range of examples, some of which are very new (e.g., bioinformatics) and non-standard. It contains many interesting ideas, and is concise mathematically and easy to read. Highly recommended.

I am a faculty member at Dept. of Educational & Counseling Psychology and teaches Statistics for graduate students. I owned several Multidimensional Scaling (MDS) books since I have been using MDS a lot for my own research. I read this book a couple of times for my own research. There are several outstanding aspects compared to other MDS books (e.g., books by Cox & Cox; by Borg and Groenen; and by Young): (1) Davison's MDS book is easy to read, not that much technical and easy to understand the core concepts of MDS; (2) Examples are appropriate. His illustrations really help understand what he covered in each chapter; (3) Davison's comparions of MDS with its alternatives are very meaningful, such as comparions with Factor anaysis and Cluster analysis; (4) Davison's exercises sections are really helpful to make sure whether the users understand materials covered in each chapter. In addition to that, the exercises's solutions for each chapter were attached at the end of each chapter, which is really convenient for us.

In summary, as far as I am concerned, Davison's book is the best for introductory or advanced level students who want to learn about MDS. Excellent book! If you have any further questions regarding Davison's MDS book, please don't hesitate call or contact me: e-mail is 'kimse@missouri.edu' or phone is '573-884-5587.' Thanks.

Se-Kang Kim, Ph.D. 16 Hill Hall, Dept. of Educational & Counseling Psychology, Univ. of Missouri

Columbia, MO 65211

Rupert Miller's book "Simultaneous Statistical Inference" was acclaimed when it was published in 1966 as a source that unified the scattered literature on the topic. In 1981, the second edition provided a welcomed update.

Since that time the methods and applications continue to grow. Hsu provides a systematic treatment and makes the material accessible. The literature is often disjoint and confusing but Hsu puts things together very well and provides a wealth of applications in a variety of disciplines including pharmaceuticals and quality control engineering.

The book is also very valuable because it exposes the misuses. It is probably the best current reference source for this subject.

This book gives the clearest and most elegant presentation of the theory of multivariate analysis I have seen. The reader should have a good background in linear algebra before starting this one, but with this background the authors give a very concise treatment of a large area of statistics. Many topics that are not covered in most multivariate analysis texts are covered here. The exercises are also outstanding. Many of them are very difficult and lead to more profound results. I highly recommend this text!

This is a great source for anyone who wishes to learn about some exciting new developments in multivariate approximation theory: A lovely book with a good mix of numerical and theoretic work. The individual contributions are well written, attractive, and by authorities in the field. A sample of subjects: Applications of scaling theory to subdivision algorithms, interpolation, radial bases, computer graphics, non-uniform splines...