Computer-Aided Multivariate Analysis

This book has been written for investigators, specifically behavioral scientists, biomedical scientists, industrial or academic researchers who wish to perform multivariate statistical analyses on their data and understand the results. lt has been written so that it can either be used as a self-guided textbook or as a text in an applied course in multivariate analysis. In addition, we believe that the book will be helpful to many statisticians who have been trained in conventional mathematical statistics who are now working as statistical consultants and need to give explanations to clients who Iack sufficient background in mathematics. We do not present mathematical derivations ofthe techniques in this book; rather we rely on geometric and graphical arguments and on examples to illustrate them. The mathematicallevel has been kept deliberately low, with no mathematics beyond the high-school Ievel required. The derivations of the techniques are referenced. The original derivations for most of the current techniques were done 50 years ago so we feel that the applications of these techniques to real-life problems is the 'fun' part now. We have assumed that the reader has taken a basic course in statistics that includes test of hypotheses. Many computer programs use analysis of variance and that part of the results of the program can only be understood if the reader is familiar with one-way analysis of variance.