Statistique Théorique et Appliquée by P. DagnelieReview by: Naguib LallmahomedJournal of the Royal Statistical Society. Series A (Statistics in Society), Vol. 170, No. 4 (2007),p. 1180Published by: Wiley for the Royal Statistical SocietyStable URL: http://www.jstor.org/stable/4623233 .

Accessed: 28/06/2014 10:37

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

Wiley and Royal Statistical Society are collaborating with JSTOR to digitize, preserve and extend access toJournal of the Royal Statistical Society. Series A (Statistics in Society).

http://www.jstor.org

This content downloaded from 91.213.220.103 on Sat, 28 Jun 2014 10:37:46 AMAll use subject to JSTOR Terms and Conditions

1180 Book Reviews

highly mathematical and would not be easy reading for someone without a good grounding and under- standing of statistical theory. Although I think that the authors have achieved their aim of providing a

'complete, comprehensive, and updated reference and textbook in the area'

primarily for the statistics student, it would have been good if the theory had been better supported by discussion of context and application. For exam- ple, I am not convinced of the rationale for adap- tive hypotheses or whether it is reasonable ever to let data that are collected early in the trial alter the 'clinically meaningful difference'. These aspects are skirted over and although the equations that are applied are valid for the purpose I would have liked to be more convinced of the validity of that purpose. The final chapter offers a useful overview of points to note when designing an adaptive trial and also some case-studies of the methods that are explained in earlier chapters. It is worth alerting the reader to the existence of these examples to refer to while reading the preceding chapters as they may assist understanding.

Non-mathematical users will require support in understanding the formulae and their interpreta- tions presented as there are important omissions which suppose that the reader has a more than basic level of knowledge. For example, terms like 'eigen- value' are introduced without explanation and then used to compare randomization methods without explanation of how they should be interpreted in this context. The book should not be read in isola- tion as it is desirable that any users of these meth- ods have a greater understanding of the ethical and practical issues than are presented. In summary, this book is worth reading for someone with a statisti- cal background who wants a general overview of current statistical methodologies.

Angela Wade University College London

Statistique Theorique et Appliquee, vol. II P. DAGNELIE, 2006 Brussels, De Boeck and Larcier 734 pp., C39 ISBN 978-2-804-15229-1

This book by Professor Dagnelie (a former Pres- ident and Honorary Life Member of the Interna- tional Biometric Society and an Honorary Fellow of the Royal Statistical Society) is a revamped edition of his earlier 1998 book, with considerable additions of new material.

Whereas the first volume of this series is mainly dedicated to undergraduates of French-speaking

universities, this second updated volume aims mostly at advanced undergraduates and beginning graduates and lecturers.

Professor Dagnelie's book also has the advantage that it can be used as a classical reference guide in major research centres, particularly those dealing with the biological and agricultural sciences.

One important feature of this book is that the text is clear and accessible for beginners and would- be specialists alike. It provides numerous worked examples and exercises, all of which help to build the necessary statistical skills that are necessary for the training of a good statistician. Answers to the exer- cises that are set are provided at the end of the book.

Another feature of the book is its bibliography, which contains more than 900 entries covering 46 pages. Various errors which appeared in the previ- ous 1998 edition have also been corrected.

Reviewing an earlier edition in the Journal of the Royal Statistical Society, Series A, Professor D. J. Finney had this to say:

'The book appears to be very well designed, either as a text for a formal course, or for refer- ence, it will give a good appreciation of statistical methodology'.

The only thing which we might regret about Pro- fessor Dagnelie's book is that it is written for a French-speaking audience. I firmly believe that the English version of such a statistical reference guide (by someone who has taught statistics for more than 35 years) should be widely acclaimed.

Naguib Lallmahomed Mauritius

Statistical Analysis of Medical Data using SAS G. DER AND B. S. EVERITT, 2006 Boca Raton, Chapman and Hall 428 pp., ?39.99 ISBN 1-584-88469-X

This is a book that I wish that I had had 15 years ago when I first started using SAS, but I am still glad to have it now. It starts with a basic introduc- tion to the SAS user interface and introduces the basics of SAS programming, reading in and manip- ulating data sets and writing basic macros, and then moves on to basic graphs and the output delivery system. It covers the topics that you would expect to see in any basic medical statistics text-book such as basic inference and regression, but it also includes topics such as generalized additive models, general- ized estimating equations and survival analysis with time varying covariates. The text is well laid out and easy to read, with numerous practical examples that

This content downloaded from 91.213.220.103 on Sat, 28 Jun 2014 10:37:46 AMAll use subject to JSTOR Terms and Conditions