The linear discriminant function; Discrete classification models; Error rates and the problem of bias; The variable-selection problem; Special topics; Computer programs.

When predicting group membership, success on an external criterion, mastery in a particular subject area, etc., there are of course many discriminant analysis procedures that might be applied. In many cases the techniques considered are a function of the type of data that is available. The goal in this paper is to discuss some of the author's recent work that is relevant to discriminant analysis, comment on recent related investigations by other investigators, and suggest directions for future research.

The Wiley-Interscience Paperback Series consists of selected books that have been made more accessible to consumers in an effort to increase global appeal and general circulation. With these new unabridged softcover volumes, Wiley hopes to extend the lives of these works by making them available to future generations of statisticians, mathematicians, and scientists. "For both applied and theoretical statisticians as well as investigators working in the many areas in which relevant use can be made of discriminant techniques, this monograph provides a modern, comprehensive, and systematic account of discriminant analysis, with the focus on the more recent advances in the field." –SciTech Book News ". . . a very useful source of information for any researcher working in discriminant analysis and pattern recognition." –Computational Statistics Discriminant Analysis and Statistical Pattern Recognition provides a systematic account of the subject. While the focus is on practical considerations, both theoretical and practical issues are explored. Among the advances covered are regularized discriminant analysis and bootstrap-based assessment of the performance of a sample-based discriminant rule, and extensions of discriminant analysis motivated by problems in statistical image analysis. The accompanying bibliography contains over 1,200 references.

Often the scientist is faced with a large number of categorical variates which are of potential use in discriminating between two pregiven groups of objects. For example, an investor may wish to assign a particular firm to one of two possible risk groups based upon certain known characteristics of the firm (liquid to fixed asset ratio, etc.), or an engineer might wish to determine which of two models best describes a particular situation based upon the observed characteristics of situation. This is the general problem of variable selection in discriminant analysis. When obtaining and processing the numerous variables is expensive, one must select a best subset of variables which incorporates as much information for discriminating as possible. If time is also a factor, a stepwise procedure is mandated. We propose such a stepwise procedure here based upon information theoretic considerations. (Author).