Review of simple regression. Introduction to matrices. Multiple regression in matrix notation. Analysis of variance and quadratic forms. Case study: five independent variables. Geometric interpretation of least squares. Model development: selection of variables. Class variables in regression. Problem areas in least squares. Regression diagnostics. Transformations of variables. Colinearity. Case study: colinearity problems. Response curve modeling. Case study: response surface modeling. Analysis of unbalanced data. Case study: analysis of unbalanced data. Appendix. Tables. Answers to selected exercises. Bibliography. Index.
(source: Nielsen Book Data)
Providing an introduction to modern regression analysis, this text emphasises understanding concepts rather than computational details or mathematical proofs. The frequent discussions of computing with SAS allows emphasis on whole concepts rather than conceptual details. This text includes topics unique for a book at this level, such as a chapter on a geometric interpretation of least square analysis, coverage of reparametrization, regression diagnostics including Gabriel's biplots and the analysis of unbalanced data. The text uses matrix notation throughout and includes extensive discussions on the application of regression models to analysis of variance problems, the limitations of stepwise regression procedures and of analyzing highly colinear data. This book should be of interest to degree and diploma students taking courses in statistics. (source: Nielsen Book Data)