Multivariate Statistical Analysis: Old School 2

A thorough introduction to multivariate statistical analysis, covering the multivariate normal distribution, linear models (MANOVA, GMANOVA, multivariate regression, growth curves), covariance models (factor analysis, invariant normal models, independence and conditional independence), classification, clustering, principal components, canonical correlations, and multidimensional scaling. The approach is old school, i.e., mathematically rigorous, although presented in such a way that the reader can learn the basic methodology while delving into the justification as much or as little as desired.