This text is about ideas for the graphical analysis of regression data. A statistical context for regression and regression graphics is emphasized. This new context is intended to blend with, rather than replace, most traditional paradigms for regression analysis. Residuals are an important part of graphics for regression analysis, and they play key roles in this book. But they are not singled out for special study; rather, they occur throughout the volume in different roles depending on the context.

Regression analysis is the study of the dependence of a response variable on one or more predictor variables. It is among the most widely used methods in statistics. There are several ways to approach regression. First, the types of models that might be used have been expanded from the traditional linear and nonlinear regression models to generalised linear models (GLMs) and others. Second, non- and semi-parametric approaches have been tried, but these have as yet only a limited impact on the general practice of statistics. Finally, computer graphics, including high quality images, motion,...

Written by the leading expert in the field, this text reviews the major new developments in envelope models and methods An Introduction to Envelopes provides an overview of the theory and methods of envelopes, a class of procedures for increasing efficiency in multivariate analyses without altering traditional objectives.