Recursive Partitioning in the Health Sciences

1. Front Matter

   Pages i-xii

   PDF

2. Introduction

   • Heping Zhang, Burton Singer

   Pages 1-6

3. A Practical Guide to Tree Construction

   • Heping Zhang, Burton Singer

   Pages 7-19

4. Logistic Regression

   • Heping Zhang, Burton Singer

   Pages 21-27

5. Classification Trees for a Binary Response

   • Heping Zhang, Burton Singer

   Pages 29-59

6. Risk-Factor Analysis Using Tree-Based Stratification

   • Heping Zhang, Burton Singer

   Pages 61-69

7. Analysis of Censored Data: Examples

   • Heping Zhang, Burton Singer

   Pages 71-77

8. Analysis of Censored Data: Concepts and Classical Methods

   • Heping Zhang, Burton Singer

   Pages 79-92

9. Analysis of Censored Data: Survival Trees

   • Heping Zhang, Burton Singer

   Pages 93-103

10. Regression Trees and Adaptive Splines for a Continuous Response

    • Heping Zhang, Burton Singer

    Pages 105-135

11. Analysis of Longitudinal Data

    • Heping Zhang, Burton Singer

    Pages 137-172

12. Analysis of Multiple Discrete Responses

    • Heping Zhang, Burton Singer

    Pages 173-199

13. Appendix

    • Heping Zhang, Burton Singer

    Pages 201-209

14. Back Matter

    Pages 211-226

    PDF

Multiple complex pathways, characterized by interrelated events and con­ ditions, represent routes to many illnesses, diseases, and ultimately death. Although there are substantial data and plausibility arguments supporting many conditions as contributory components of pathways to illness and disease end points, we have, historically, lacked an effective methodology for identifying the structure of the full pathways. Regression methods, with strong linearity assumptions and data-based constraints on the extent and order of interaction terms, have traditionally been the strategies of choice for relating outcomes to potentially complex explanatory pathways. How­ ever, nonlinear relationships among candidate explanatory variables are a generic feature that must be dealt with in any characterization of how health outcomes come about. Thus, the purpose of this book is to demon­ strate the effectiveness of a relatively recently developed methodology­ recursive partitioning-as a response to this challenge. We also compare and contrast what is learned via recursive partitioning with results ob­ tained on the same data sets using more traditional methods. This serves to highlight exactly where--and for what kinds of questions-recursive partitioning-based strategies have a decisive advantage over classical re­ gression techniques. This book is suitable for three broad groups of readers: (1) biomedical re­ searchers, clinicians, public health practitioners including epidemiologists, health service researchers, environmental policy advisers; (2) consulting statisticians who can use the recursive partitioning technique as a guide in providing effective and insightful solutions to clients' problems; and (3) statisticians interested in methodological and theoretical issues.

• Computerassistierte Detektion
• Factor analysis
• Logistic Regression
• Radiologieinformationssystem
• Recursive Partitioning
• Statistics for Health Sciences
• classification
• methodology
• statistics

STATISTICAL METHODS IN MEDICAL RESEARCH

"The beauty of the Zhang and Singer’s book is that it gives an excellent comparison between conventional regression models and recursive partitioning techniques. This comparative approach gives the reader insight into how a recursive partitioning technique can have an advantage over the conventional methods…Overall, the book provides an excellent introduction to tree based methods and their applications. It can be a good place to start learning about recursive partitioning. In addition, biostatisticians will enjoy the real life examples that have been used in the book."

• Department of Epidemiology and Public Health School of Medicine, Yale University, New Haven, USA

  Heping Zhang

• Office of Population Research, Princeton University, Princeton, USA

  Burton Singer

Policies and ethics