Bendix Carstensen, Senior Statistician,
Clinical Epidemiology, Steno Diabetes Center, Gentofte, Denmark, &
Department of Biostatistics, Institute of Public Health, University of Copenhagen, Denmark
- Here are a few useful program chuncks --- essentially the solution code for the practicals.
- Bring your own computer with the latest version of R (3.0.2) installed.
- Install the latest version of the MethComp package (1.27). This is not (yet) on CRAN, but available as .zip here.
- It will be useful to install the Epi package too.
- Finally, in order to run the MCMC simulations from the MethComp package, you must install JAGS on your computer.
- Here is a small R program that should run on your computer if all is installed properly.
There might be possibilities of looking at your own data. It would be preferable if you contacted the teacher first to discuss if it is an example worth including in the course.
- Papers:
- M Bland & DG Altman: Statistical methods for assessing agreement between two methods of clinical measurement.
- M Bland & DG Altman: Measuring agreement in method comparison studies.
- B Carstensen, JA Simpson & L Gurrin: Statistical models for assessing agreement in method comparison studies with replicate measurements.
- B Carstensen: Comparing methods of measurement: Extending the LoA by regression.
- Book: Bendix Carstensen: Comparing Clinical Measurement Methods, Wiley, 2010, ISBN 978-0-470-69423-7.
