## Welcome to Cengage

Would you like to be redirected to the site for United States?

Go to the site for United States

Stay on current Cengage siteWould you like to be redirected to the site for United States?

Go to the site for United States

Stay on current Cengage sitePrint £86.99

Using a data-driven approach, this book is an exciting blend of theory and interesting regression applications. Students learn the theory behind regression while actively applying it. Working with many case studies, projects, and exercises from areas such as engineering, business, social sciences, and the physical sciences, students discover the purpose of regression and learn how, when, and where regression models work. The book covers the analysis of observational data as well as of data that arise from designed experiments. Special emphasis is given to the difficulties when working with observational data, such as problems arising from multicollinearity and "messy" data situations that violate some of the usual regression assumptions. Throughout the text, students learn regression modeling by solving exercises that emphasize theoretical concepts, by analyzing real data sets, and by working on projects that require them to identify a problem of interest and collect data that are relevant to the problem's solution. The book goes beyond linear regression by covering nonlinear models, regression models with time series errors, and logistic and Poisson regression models.

1. Introduction to Regression Models.

2. Simple Linear Regression.

3. A Review of Matrix Algebra and Important Results of Random Vectors.

4. Multiple Linear Regression Model.

5. Specification Issues in Regression Models.

6. Model Checking.

7. Model Selection.

8. Case Studies in Linear Regression.

9. Nonlinear Regression Models.

10. Regression Models for Time Series Situations.

11. Logistic Regression.

12. Generalized Linear Models and Poisson Regression.

Brief Answers to Selected Exercises.

Statistical Tables.

References.

- Case studies throughout the text illustrate the process of regression modeling and emphasize its benefits, while also warning of its pitfalls and problems.
- In addition to teaching the use of regression, the text also provides rigorous coverage of the theory behind regression. This gives students the theoretical foundation that is needed for subsequent courses and further self-study. The more theoretical portions of the book can be omitted without compromising the main ideas.
- Chapter 1 addresses in detail the purpose of regression to prepare students for the course.
- Projects in the book address questions that interest readers from diverse fields of study. The wide variety includes topics such as future U.S. presidential elections, fuel efficiencies of automobiles, the role of race in death penalty sentencing, scholastic achievement of U.S. students, French wine prices, sales effects of advertising, the survival of the Donner party, and many more.
- Chapter 8 provides suggestions for project topics and guidelines for dealing with projects. Successful projects involve the application of the studied regression techniques to solve real problems.
- The book goes beyond the typical linear regression model by covering nonlinear models, regression models with time series errors, and logistic and Poisson regression models. These topics are important since the response data in many application areas are categorical involving counts, and because observations often arise in the form of time series.
- Many of the exercises are based on real situations and give students ample chance to practice on meaningful problems. The exercises were pre-tested in various classes on regression, time series modeling (Chapter 10), and statistical methods for business applications (Chapter 11 on logistic regression).
- The text is not tied to a specific computer program, and it discusses computer output from several commonly used packages such as MINITAB®, R®, S-Plus®, SPSS®, and SAS®.
- The data-driven approach teaches practical modeling skills. Students learn how, when, and where regression models work.

**Bovas Abraham**

University of Waterloo

Bovas Abraham is the former Director of the Institute for Improvement in Quality and Productivity, and is also a professor in the Department of Statistics and Actuarial Science at the University of Waterloo. Bovas received his Ph.D. from the University of Wisconsin, Madison. He has held visiting positions at the University of Wisconsin, the University of Iowa, and the University of Western Australia. He is the author of the book STATISTICAL METHODS FOR FORECASTING (with Johannes Ledolter) published by Wiley in 1983, and the editor of the volume QUALITY IMPROVEMENT THROUGH STATISTICAL METHODS published by Birkhauser in 1998.

**Johannes Ledolter**

University of Iowa

Johannes Ledolter is the John F. Murray Professor of Management Sciences at the University of Iowa, and a Professor at the Vienna University of Economics and Business Administration. His graduate degrees are in Statistics (M.S. and Ph.D. from the University of Wisconsin, and M.S. from the University of Vienna). He has held visiting positions at Princeton University and Yale University. He is the author of four books: STATISTICAL METHODS FOR FORECASTING (with Bovas Abraham) published by Wiley in 1983, STATISTICS FOR ENGINEERS AND PHYSICAL SCIENTISTS (2nd edition, with Robert V. Hogg) published by Macmillan in 1991, STATISTICAL QUALITY CONTROL (with Claude W. Burrill) published by Wiley in 1999, and ACHIEVING QUALITY THROUGH CONTINUAL IMPROVEMENT (with Claude W. Burrill) published by Wiley in 1999.