Multivariable model-building
Multivariable model-building
A pragmatic approach to regression analysis based on fractional polynomials for modelling continuous variables
A two-day course by Prof Patrick Royston (London) & Prof Willi Sauerbrei (Freiburg)
Course aims:-
* To appreciate the issues surrounding the development of a multivariable model in observational studies
* To learn about strategies for selecting variables in such a model
* To learn about the fractional polynomials method for modelling continuous predictors
* To be introduced to multivariable model-building using the multivariable fractional polynomials (MFP) method
* To understand methods for modelling interactions in randomized trials and observational studies
* To understand issues around model (in)stability
* On issues in spline modeling and comparison to MFP in examples
* To learn how to work in a practical way with MFP
We invite participants to submit their own data and problems, but no later than THREE WEEKS before the start of the course, i.e. the deadline is THURSDAY 21st JANUARY 2010. One session is dedicated to discussion and analysis of some participants' data. Problems are limited to the course aims and exclude topics such as measurement error, multi-level models, longitudinal data (other than survival times) or missing covariate observations.
The course follows closely the book with the same title by the lecturers, published by Wiley (2008). A copy of the book is included in the course material.
Nearly all datasets and many programs in Stata can be downloaded from the book's website: http://www.imbi.uni-freiburg.de/biom/Royston-Sauerbrei-book/
Target audience:-
* Practical statisticians and researchers interested in extending their knowledge about multivariable regression model building and handling of continuous predictors in particular
* Some experience of real data analysis using linear, logistic and Cox regression is assumed
Lecturers:-
* Patrick Royston, MRC Clinical Trials Unit, UK
* Willi Sauerbrei, Institute of Medical Biometry and Medical Informatics, University Medical Center Freiburg
To register for the course please use the link below - should you have any problems please contact Kathy Airey - contact details below.
www.mrc-bsu.cam.ac.uk/Education/BSUcourses.html
Kathleen Airey, PA / Assistant Administrator
MRC-Biostatistics Unit
IPH Forvie Site, Robinson Way
Cambridge, CB2 0SR
Tel: 01223 330366 - Direct Line
01223 330376 - second line
Fax: 01223 330365
Email: kathy.airey@mrc-bsu.cam.ac.uk
