You are here

Modern Methods for Robust Regression
Share

Modern Methods for Robust Regression



© 2008 | 128 pages | SAGE Publications, Inc

Modern Methods for Robust Regression offers a brief but in-depth treatment of various methods for detecting and properly handling influential cases in regression analysis. This volume, geared toward both future and practicing social scientists, is unique in that it takes an applied approach and offers readers empirical examples to illustrate key concepts. It is ideal for readers who are interested in the issues related to outliers and influential cases.

Key Features

  • Defines key terms necessary to understanding the robustness of an estimator: Because they form the basis of robust regression techniques, the book also deals with various measures of location and scale.
  • Addresses the robustness of validity and efficiency: After having described the robustness of validity for an estimator, the author discusses its efficiency.
  • Focuses on the impact of outliers: The book compares the robustness of a wide variety of estimators that attempt to limit the influence of unusual observations.
  • Gives an overview of some traditional techniques: Both formal statistical tests and graphical methods detect influential cases in the general linear model.
  • Offers a Web appendix: This volume provides readers with the data and the R code for the examples used in the book.

Intended Audience

This is an excellent text for intermediate and advanced Quantitative Methods and Statistics courses offered at the graduate level across the social sciences.

Learn more about "The Little Green Book" - QASS Series! Click Here

 
List of Figures
 
List of Tables
 
Series Editor's Introduction
 
Acknowledgments
 
1. Introduction
Defining Robustness  
Defining Robust Regression  
A Real-World Example: Coital Frequency of Married Couples in the 1970s  
 
2. Important Background
Bias and Consistency  
Breakdown Point  
Influence Function  
Relative Efficiency  
Measures of Location  
Measures of Scale  
M-Estimation  
Comparing Various Estimates  
Notes  
 
3. Robustness, Resistance, and Ordinary Least Squares Regression
Ordinary Least Squares Regression  
Implications of Unusual Cases for OLS Estimates and Standard Errors  
Detecting Problematic Observations in OLS Regression  
Notes  
 
4. Robust Regression for the Linear Model
L-Estimators  
R-Estimators  
M-Estimators  
GM-Estimators  
S-Estimators  
Generalized S-Estimators  
MM-Estimators  
Comparing the Various Estimators  
Diagnostics Revisited: Robust Regression-Related Methods for Detecting Outliers  
Notes  
 
5. Standard Errors for Robust Regression
Asymptotic Standard Errors for Robust Regression Estimators  
Bootstrapped Standard Errors  
Notes  
 
6. Influential Cases in Generalized Linear Models
The Generalized Linear Model  
Detecting Unusual Cases in Generalized Linear Models  
Robust Generalized Linear Models  
Notes  
 
7. Conclusions
 
Appendix: Software Considerations for Robust Regression
 
References
 
Index
 
About the Author
Key features

 

  • This volume offers applied coverage of a topic that has traditionally been discussed from a theoretical standpoint.
  • The authors uses empirical examples to illustrate key concepts,
  • A Web Appendix provides readers with the data and the R-code for the examples used in the book.

 


Sample Materials & Chapters

Chapter 2

Chapter 4

Chapter 6

Andersen_files.zip


Preview this book

For instructors

This book is not available as a review copy. For more information contact your local sales representative.

Purchasing options

Please select a format:

ISBN: 9781412940726

This title is also available on SAGE Research Methods, the ultimate digital methods library. If your library doesn’t have access, ask your librarian to start a trial.