Applied multivariate analysis / Neil H. Timm.

Includes bibliographical references (pages 625-666) and indexes.

Cover -- Preface -- Acknowledgments -- Contents -- List of Tables -- List of Figures -- 1. Introduction -- 2. Vectors and Matrices -- 3. Multivariate Distributions and the LInear Model -- 4. Multivariate Regression Models -- 5. Seemingly Unrelated Regression Models -- 6. Multivariate Random and MIxed Models -- 7. Discriminant and Classification Analysis -- 8. Principal Component, Canonical Correlation, and Exploratory FActor Analysis -- 9. Cluster Analysis and Multidimensional Scaling -- 10. Structural Equation Models -- Appendix A -- References -- Author Index.

This textbook provides a broad overview of the basic theory and methods of applied multivariate analysis. The presentation integrates both theory and practice including both the analysis of formal linear multivariate models and exploratory data analysis techniques. Each chapter contains the development of basic theoretical results with numerous applications illustrated using examples from the social and behavioral sciences, and other disciplines. All examples are analyzed using SAS for Windows Version 8.0. The book includes an overview of vectors, matrices, multivariate distribution theory, and multivariate linear models. Topics discussed include multivariate regression, multivariate analysis of variance for fixed and mixed models, seemingly unrelated regression models and repeated measurement models. While standard procedures for estimating model parameters and testing multivariate hypotheses, as well as simultaneous test procedures, are discussed and illustrated in the text, the text also includes tests of multivariate normality with chi-square and beta plots, tests of multivariate nonadditivity, tests of covariance structure, tests of nonnested hypotheses, and the assessment of model assumptions. Other topics illustrated in the text include discriminant and classification analysis, principal component analysis, canonical correlation analysis, exploratory factor analysis, cluster analysis, multidimension scaling, and structural equation modeling. The text should appeal to practitioners, researchers, and applied statisticians. It may be used in a one-semester course in applied multivariate analysis for practitioners and researchers, or as a two- semester course for majors in applied statistics. Because most data analyzed in the social and behavioral sciences and other disciplines involve many continuous variables, the techniques and examples.