Model selection and inference. A practical information theoretic approach. (English) Zbl 0920.62006

Burnham, Kenneth P.; Anderson, David A.

New York, NY: Springer. xix, 353 p. DM 149,00; öS 1.088,00; sFr 136,00; £57,50; $ 69,95 (1998).

This book covers some philosophy about data analysis, some theory at the interface between mathematical statistics and information theory, and some practical statistical methodology useful in the life sciences. The book is written to introduce graduate students and research workers in various scientific disciplines to the use of information-theoretic approaches in the analysis of empirical data. The main focus is on Akaikes information criterion (and various extensions) for selection of a parsimonious model as a basis for statistical inference.
Chapters 1-3 are somewhat tutorial, while Chapters 4 and 5 present new research results and a wide variety of approaches illustrated with examples. Chapter 6 covers much more advanced mathematical material, while a comprehensive summary of the book is in Chapter 7. Finally, the book gives references from the diverse literature on these subjects for those interested in further studies. The book will be useful to biologists and statisticians using models for making inferences from empirical data.

Reviewer: V.P.Gupta (Jaipur)

MSC:

62B10	Statistical aspects of information-theoretic topics
62-01	Introductory exposition (textbooks, tutorial papers, etc.) pertaining to statistics
62-02	Research exposition (monographs, survey articles) pertaining to statistics