Methods of multivariate analysis

About this book

Analyzing data from experiments that yield large numbers of intercorrelated variables can be a daunting task. With hundreds or even thousands of separate numbers to inspect, analysts use special tools known as multivariate statistical methods to locate and identify latent patterns within the raw data. This comprehensive volume - the finest introduction to the subject available - covers the most reliable multivariate techniques and offers many insights that can otherwise be found only in journal articles or in the minds of practitioners. Developed by Professor Alvin C. Rencher from his one-semester course at Brigham Young University, this book is tailored to the needs of students who are getting their first exposure to multivariate analysis. The careful, intuitive explanations of concepts and procedures are a model of clarity, and simple proofs provide a solid grounding for statistics majors while remaining accessible to nonmajors as well. Since many multivariate techniques are extensions of analogous univariate methods, a careful review of these univariate procedures is also provided. The appendixes supply tables for many multivariate distributions and tests, together with answers and hints for the problems. . Methods of Multivariate Analysis also contains more applied problems involving real data sets than any other text. The 41 data sets are taken from a broad range of applications, including: agriculture, biology, business, dentistry, economics, engineering, medicine, pharmacology, psychology, literature, sports, and education. The accompanying diskette contains all of the data sets and SAS command files for all of the examples. (SAS is the leading statistical computer package on the market.) Students can adapt these command files to work problems in the text. Offering broader, more detailed, and more accessible coverage than other books in the field, and demanding a lower level of required mathematics than competing volumes, Methods of Mulivariate Analysis is at once the best general purpose multivariate methods text for statistics courses and an invaluable reference and practical aid to scientists, engineers, and social scientists.