|Cook, R. Dennis / Weisberg, Sanford|
Applied Regression Including Computing and Graphics
Wiley Series in Probability and Statistics
1. Edition August 1999
1999. 632 Pages, Hardcover
ISBN 978-0-471-31711-1 - John Wiley & Sons
E-Books are also available on all known E-Book shops.
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. In recent years, several new ways to approach regression have been presented. First, the types of models that might be used have been expanded from the traditional linear and nonlinear regression models to generalized 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, color, plot linking, and so on, have been applied to the regression problem. This book ties together all these modern trends in regression analysis using graphics as a unifying theme. The book present the essential theory and practice based on standard models and on selected non- and semi-parametric ideas.
From the contents
Looking Forward and Back.
Introduction to Regression.
Introduction to Smoothing.
Simple Linear Regression.
Introduction to Multiple Linear Regression.
Weights and Lack-of-Fit.
Relating Mean Functions.
Factors and Interactions.
Diagnostics I: Curvature and Nonconstant Variance.
Diagnostics II: Influence and Outliers.
Visualizing Regression with Many Predictors.
LOGISTIC REGRESSION AND GENERALIZED LINEAR MODELS.
Graphical and Diagnostic Methods for Logistic Regression.
Generalized Linear Models.