Multivariable Model-Building:
A pragmatic approach to regression analysis based on fractional polynomials for modelling continuous variables



About this Book
Multivariable regression models are of fundamental importance in all areas of science in which empirical data must be analyzed. This book proposes a systematic approach to building such models based on standard principles of statistical modeling. The main emphasis is on the fractional polynomial method for modeling the influence of continuous variables in a multivariable context, a topic for which there is no standard approach. Existing options range from very simple step functions to highly complex adaptive methods such as multivariate splines with many knots and penalisation. This new approach, developed in part by the authors over the last decade, is a compromise which promotes interpretable, comprehensible and transportable models.



- Contents

- Preface

1 Introduction

2 Selection of Variables

3 Handling Categorical and Continuous Predictors

4 Fractional Polynomials for One Variable

5 Some Issues with Univariate Fractional Polynomial Models

6 Multivariable Fractional Polynomial: Multivariable Model-Building with Fractional Polynomials

7 Interactions

- Interaction

8 Model Stability

9 Some Comparisons of Multivariable Fractional Polynomials with Splines

10 How To Work with Multivariable Fractional Polynomials

11 Special Topics Involving Fractional Polynomials

12 Epilogue

13 Data and Software Resources

- Appendix: Glossary of Abbreviations

- References

[size=1.2em]Author(s): Patrick Royston, Willi Sauerbrei

[size=1.2em]Published Online: 7 MAY 2008

[size=1.2em]Print ISBN: 9780470028421

[size=1.2em]Online ISBN: 9780470770771

[size=1.2em]DOI: 10.1002/9780470770771

[size=1.2em]Book Series: [size=1em]Wiley Series in Probability and Statistics

[size=1.2em]Series Editor(s): Walter A. Shewhart, Samuel S. Wilks