zbMATH — the first resource for mathematics

The statistical analysis of discrete data. (English) Zbl 0702.62005
Springer Texts in Statistics. New York etc.: Springer-Verlag. xii, 367 p. DM 88.00 (1989).
The book deals with discrete data and covers adequately methods and techniques to analyse univariate, polychotomous and multinomial data. In the first chapter, the authors introduce the inquisitive reader to a variety of problems yielding discrete data in various disciplines. Chapter 2 covers widely the Bayes, hierarchical, empirical and pseudo- Bayes estimators. Point and interval estimation, testing of hypotheses with global and isotonic alternatives are covered in a logical manner including latest developments.
Chapter 3 deals with loglinear models in a lucid manner. Chapter 4 elucidates multidimensional cross classification tables and hierarchical, graphical and discrete linear models. Chapter 5 gives methods for estimating parameters of polychotomous response regression and logistic regression. It also contains methods for finding simultaneous confidence intervals for all pairwise differences. The applications of the theory are demonstrated very well through real data problems mostly from medical sciences.
On the whole it is a well written compendium for graduate students and also providing readymade material to the applied scientist.
Reviewer: B.L.Agarwal

62-02 Research exposition (monographs, survey articles) pertaining to statistics
62F99 Parametric inference
62F03 Parametric hypothesis testing
62F10 Point estimation
62F25 Parametric tolerance and confidence regions
AS 129; AS 159