Handbook of Regression Methods

Handbook of Regression Methods concisely covers numerous traditional, contemporary, and nonstandard regression methods. The handbook provides a broad overview of regression models, diagnostic procedures, and inference procedures, with emphasis on how these methods are applied. The organization of the handbook benefits both practitioners and researchers, who seek either to obtain a quick understanding of regression methods for specialized problems or to expand their own breadth of knowledge of regression topics.

This handbook covers classic material about simple linear regression and multiple linear regression, including assumptions, effective visualizations, and inference procedures. It presents an overview of advanced diagnostic tests, remedial strategies, and model selection procedures. Finally, many chapters are devoted to a diverse range of topics, including censored regression, nonlinear regression, generalized linear models, and semiparametric regression.

Features

• Presents a concise overview of a wide range of regression topics not usually covered in a single text
• Includes over 80 examples using nearly 70 real datasets, with results obtained using R
• Offers a Shiny app containing all examples, thus allowing access to the source code and the ability to interact with the analyses

Introduction. Simple Linear Regression. The Basics of Regression Models. Statistical Inference. Statistical Intervals. Assessing Regression Assumptions. ANOVA I. Multiple Linear Regression. Multiple Regression. Matrix Notation in Regression. Indicator Variables. Multicollinearity. ANOVA II. Advanced Regression Diagnostic Methods. Influential Data Values. Measurement Errors and Instrumental Variables Regression. Weighted Least Squares and Robust Regression Procedures. Correlated Errors and Autoregressive Structures. Crossvalidation and Model Selection Methods. Advanced Regression Models. Biased Regression Methods and Regression Shrinkage. Piecewise and Nonparametric Methods. Regression Models with Censored Data. Nonlinear Regression. Regression Models with Counts as Responses. Multivariate Multiple Regression. Data Mining. Miscellaneous Topics. Appendices.

Derek Young is an assistant professor of statistics at the University of Kentucky. He has over ten years of experience as a statistician, including positions in industry, government, and academia. During this time, he has also taught online courses in regression methods for Penn State University and the University of Kentucky. His research interests include (finite) mixture models, tolerance regions, and statistical computing.