An Introduction to Regression Graphics (Wiley Series in Probability and Statistics) - Hardcover

Understanding how a response variable depends on one or more predictor variables is a universal scientific problem. Regression analysis consists of ideas and methods for addressing this problem. Historically, regression methods have been largely numerical, with graphics playing an important but subsidiary role. By allowing informative and novel visualizations of regression data, modern computer hardware and software promise to reverse the historical roles of numerical and graphical regression methods. How shall this be done in practice? What can be learned from graphs and which graphs should be drawn? How can graphs be used to learn about fundamental features of regression problems? An Introduction to Regression Graphics answers these questions and more, providing the ideas, methodology, and software needed to use graphs in regression. From simple manipulations, such as changing the aspect ratio and marking points, to more sophisticated ideas like extracting smooths or looking at uncorrelated directions in 3D plots, R. Dennis Cook and Sanford Weisberg provide step-by-step software instructions and concise explanations of how graphs can be used in almost any regression problem. The accompanying disks?compatible with Macintosh, Windows, and UNIX workstations?contain an interactive regression program called R-code along with many sample data sets and demonstrations. The R-code is capable of performing all computations discussed in the book, and can be used with your own data. The book serves as a manual for the software. For data analysts, An Introduction to Regression Graphics provides new tools and insights for graphical analysis of regression data. For graduate and upper-level undergraduate students taking regression courses in the applied sciences, this book is a source for new methods and techniques that will continue to increase in importance for years to come.

About the authors R. DENNIS COOK is Professor, Department of Applied Statistics, University of Minnesota. An active researcher in regression theory and methods, Dr. Cook is a Fellow of the American Statistical Association and of the Institute of Mathematical Statistics, and is an elected member of the International Statistical Institute. Dr. Cook received his PhD degree in statistics from Kansas State University. SANFORD WEISBERG is Professor, Department of Applied Statistics, University of Minnesota. The author of Applied Linear Regression, Second Edition (Wiley), he is a Fellow of the American Statistical Association and an elected member of the International Statistical Institute. Dr. Weisberg received his PhD degree in statistics from Harvard University. The authors have jointly published about twenty works, including Residuals and Influence in Regression and a 1989 paper on regression graphics that earned the Jack Youden prize.