Statistical Analysis of Designed Experiments

1. Front Matter

   Pages i-xv

   PDF

2. Introduction

   Pages 1-19

3. Comparison of Two Samples

   Pages 21-43

4. The Linear Regression Model

   Pages 45-110

5. Single-Factor Experiments with Fixed and Random Effects

   Pages 111-156

6. More Restrictive Designs

   Pages 157-177

7. Multifactor Experiments

   Pages 179-229

8. Models for Categorical Response Variables

   Pages 231-294

9. Repeated Measures Model

   Pages 295-340

10. Cross-Over Design

    Pages 341-384

11. Statistical Analysis of Incomplete Data

    Pages 385-413

12. Back Matter

    Pages 415-500

    PDF

This book is the second English edition of my German textbook that was originally written parallel to my lecture “Design of Experiments” which was held at the University of Munich. It is thought to be a type of resource/reference book which contains statistical methods used by - searchersin applied areas. Because of the diverse examples it could also be used in more advanced undergraduate courses, as a textbook. It is often called to our attention, by statisticians in the pharmac- tical industry, that there is a need for a summarizing and standardized representation of the design and analysis of experiments that includes the di?erent aspects of classical theory for continuous response, and of modern procedures for a categorical and, especially, correlated response, as well as more complex designs as, for example, cross–over and repeated measures. Therefore the book is useful for non statisticians who may appreciate the versatility of methods and examples, and for statisticians who will also ?nd theoretical basics and extensions. Therefore the book tries to bridge the gap between the application and theory within methods dealing with designed experiments.

• Cross over designs
• Design of experiments
• Incomplete block designs
• Linear models
• Missing data model
• linear regression

From the reviews of the second edition:

"[This book] is a useful reference or graduate text to complement more common choices for introductory design of experiment books … the methods are logically and thoroughly developed in a rigorous, yet understandable manner. The emphasis on pharmaceutical applications throughout the book is helpful, because this continues to emerge as an important area of applications. The book would be helpful for statisticians and researchers in pharmaceutical areas once they had gained a solid understanding of the fundamentals of design of experiments." –Journal of the American Statistical Association

"The second edition of this book … has been reorganized with a list of topics similar to that of the first edition, but with a revised presentation and order. … much greater emphasis now placed on the analysis aspect of design of experiments. … a useful reference book or graduate text … . The methods are logically and thoroughly developed in a rigorous, yet understandable manner. … The book would be helpful for statisticians and researchers in pharmaceutical areas … ." (Christine M. Anderson Cook, Journal of the American Statistical Association, Vol. 98 (463), 2003)

"This book is mostly concerned with the mathematical detail of the topics in the contents. There are a few sets of data, to illustrate the material; on these, SAS, S-PLUS or SPSS is used for analysis. … This would be an excellent book for mathematics students who take a course in statistics, or graduate statistic students … ." (N. R. Draper, Short Book Reviews, Vol. 23 (1), 2003)

"Helge Toutenburg describes this text as a ‘resource/reference book which contains statistical methods used by researchers in applied areas.’ … the theory is described in a shorthand style that gets to the point without overburdening the reader with mathematical detail … . the author includes thorough discussions of generalized linear models(categorical data analysis) and repeated-measures designs. … a useful, self-contained reference for those who want a quick description of the underlying theory and practice for a large assortment of standard DOE problems." (Peter Wludyka, Technometrics, Vol. 45 (2), May, 2003)

• Institut für Statistik, Universität München, München, Germany

  Helge Toutenburg

Policies and ethics