This textbook presents the design and analysis of experiments that comprises the aspects of classical theory for continuous response and of modern procedures for categorical response, and especially for correlated categorical response. For any design (independent response and matched pair response) the parametric and nonparametric tests depending on the data level are given. Complex designs, as for example, crossover and repeated measures, are included at an introductory and advanced level. The problem of missing data is discussed and the author proposes procedures for approaching this problem. This volume will be an important reference book for graduate students, university teachers, and for statistical researchers in the pharmaceutical industry and clinical research in medicine and dentistry, as well as in many other applied areas. This second edition contains more examples and graphical illustrations. Chapter 3, "The Linear Regression Model," now contains several diagnostic tools and more examples. Chapter 7, "Categorical Response Variables," was completely rewritten. The proofs of the more theoretical Chapters 3 and 4 were moved to an appendix. More emphasis has been placed on explaining and justifying some approaches. Helge Toutenburg is Professor of Statistics at the University of Munich. He has written seventeen books on linear models, statistical methods in quality engineering, and the analysis of designed experiments. He works on applications of statistics to the fields of medicine and engineering.

Скачать книгу бесплатно (pdf, 8.79 Mb)

---

Читать «Statistical Analysis Of Designed Experiments»