Linear Models for Multivariate, Time Series, and Spatial Data - Springer Texts in Statistics

This is a companion volume to the author's book "Plane Answers to Complex Questions: The Theory of Linear Models". It presents basic material on the analysis of multivariate, time series and spatial data. The exposition focuses on three ideas basic to linear model theory: best linear prediction, projection operators and Mahalanobis' distance. These are used to tie together the diverse topics into a package that requires only linear model theory as background. While frequent reference is made to "Plane Answers", all necessary material is contained in this volume. To emphasize the connections with linear model theory, much of the material is presented in a non-traditional fashion. Short discussions of more traditional approaches are also presented. The specific topics examined are multivariate linear models, discriminant analysis, principal components, factor analysis, time series in both the frequency domain and the time domain, and the universal kriging. This book is intended for graduate level students and researchers, with a background in linear model theory, who seek an introduction to any or all of the topics discussed.