zbMATH — the first resource for mathematics

Applied discriminant analysis. Incl. 1 disk. (English) Zbl 0853.62045
New York, NY: Wiley. xxiii, 466 p. (1994).
This book, containing minimum of statistical theory, focuses on applications of discriminant analysis (DA). Distinctions between using DA to predict group membership of a unit (called “predictive discriminant analysis” – Part two, chapters III to XII) and to describe group differences (“descriptive DA” – Part three, chapters XIII to XVIII) are emphasized. Developing “classification functions” for the first type of DA (multivariate rules) and “discriminant functions” for the second one (eigenanalysis) is described in detail. More considerably than in other books on DA the connection between DA and other multivariate methods (MANOVA, regression analysis, canonical correlations and so on) is presented. Problems of variable selection and ordering, missing data and outlier observations are discussed. Recommendations are given for reporting results of both predictive and descriptive DA.
The book provides numerous examples explained in great detail. At the end of each chapter a number of exercises (with answers at the end of the book) is given. An accompanying diskette contains four complete data sets and five special computer programs that are referred to in the text and are the subject of exercise problems. More than 300 references are presented.

62H30 Classification and discrimination; cluster analysis (statistical aspects)
62-01 Introductory exposition (textbooks, tutorial papers, etc.) pertaining to statistics
62-04 Software, source code, etc. for problems pertaining to statistics