This volume describes new methods with special emphasis on classification and cluster analysis. These methods are applied to problems in information retrieval, phylogeny, medical diagnosis, microarrays, and other active research areas.

Multivariate procedures allow social workers and other human services researchers to analyze complex, multidimensional social problems and interventions in ways that minimize oversimplification. This pocket guide provides a concise, practical, and economical introduction to four procedures for the analysis of multiple dependent variables: multivariate analysis of variance (MANOVA), multivariate analysis of covariance (MANCOVA), multivariate multiple regression (MMR), and structural equation modeling (SEM). Each procedure will be presented in a way that allows readers to compare and contrast them in terms of (1) appropriate research context; (2) required statistical assumptions, including levels of measurement of variables to be modeled; (3) analytical steps; (4) sample size; and (5) strengths and weaknesses. This invaluable guide facilitates course extensibility in scope and depth by allowing instructors to supplement course content with rigorous statistical procedures. Detailed annotated examples using Stata, SPSS (PASW), SAS, and Amos, together with additional resources, discussion of key terms, and a companion website, make this an unintimidating guide for producers and consumers of social work research knowledge.

A complete introduction to discriminant analysis--extensively revised, expanded, and updated This Second Edition of the classic book, Applied Discriminant Analysis, reflects and references current usage with its new title, Applied MANOVA and Discriminant Analysis. Thoroughly updated and revised, this book continues to be essential for any researcher or student needing to learn to speak, read, and write about discriminant analysis as well as develop a philosophy of empirical research and data analysis. Its thorough introduction to the application of discriminant analysis is unparalleled. Offering the most up-to-date computer applications, references, terms, and real-life research examples, the Second Edition also includes new discussions of MANOVA, descriptive discriminant analysis, and predictive discriminant analysis. Newer SAS macros are included, and graphical software with data sets and programs are provided on the book's related Web site. The book features: Detailed discussions of multivariate analysis of variance and covariance An increased number of chapter exercises along with selected answers Analyses of data obtained via a repeated measures design A new chapter on analyses related to predictive discriminant analysis Basic SPSS(r) and SAS(r) computer syntax and output integrated throughout the book Applied MANOVA and Discriminant Analysis enables the reader to become aware of various types of research questions using MANOVA and discriminant analysis; to learn the meaning of this field's concepts and terms; and to be able to design a study that uses discriminant analysis through topics such as one-factor MANOVA/DDA, assessing and describing MANOVA effects, and deleting and ordering variables.

This is the first book to compare eight LDFs by different types of datasets, such as Fisher’s iris data, medical data with collinearities, Swiss banknote data that is a linearly separable data (LSD), student pass/fail determination using student attributes, 18 pass/fail determinations using exam scores, Japanese automobile data, and six microarray datasets (the datasets) that are LSD. We developed the 100-fold cross-validation for the small sample method (Method 1) instead of the LOO method. We proposed a simple model selection procedure to choose the best model having minimum M2 and Revised IP-OLDF based on MNM criterion was found to be better than other M2s in the above datasets. We compared two statistical LDFs and six MP-based LDFs. Those were Fisher’s LDF, logistic regression, three SVMs, Revised IP-OLDF, and another two OLDFs. Only a hard-margin SVM (H-SVM) and Revised IP-OLDF could discriminate LSD theoretically (Problem 2). We solved the defect of the generalized inverse matrices (Problem 3). For more than 10 years, many researchers have struggled to analyze the microarray dataset that is LSD (Problem 5). If we call the linearly separable model "Matroska," the dataset consists of numerous smaller Matroskas in it. We develop the Matroska feature selection method (Method 2). It finds the surprising structure of the dataset that is the disjoint union of several small Matroskas. Our theory and methods reveal new facts of gene analysis.