[R-sig-ME] New cvGEE package for validating predictions from GEEs

D. Rizopoulos d@r|zopou|o@ @end|ng |rom er@@mu@mc@n|
Wed Jul 24 07:56:19 CEST 2019


Dear R mixed model users,

I’d like to announce the release of my new R package cvGEE (https://cran.r-project.org/package=cvGEE) that validates (using repeated cross-validation) predictions from generalized estimating equations using proper scoring rules. It can be used to compare nested and non-nested GEEs, such as to compare GEEs with the same mean structure but different working correlation matrices.

You may find more information and examples in the dedicated website:

https://drizopoulos.github.io/cvGEE/

and vignette

https://drizopoulos.github.io/cvGEE/articles/Scoring_Rules_GEE.html

The package currently works with geeglm objects returned by the geepack but there are plans to extend it to other GEE-solvers packages. Additional future plans include functions for calibration plots.

As always, any kind of feedback is more than welcome.

Best,
Dimitris

- - - - - -
Dimitris Rizopoulos
Professor of Biostatistics
Erasmus University Medical Center
The Netherlands

	[[alternative HTML version deleted]]



More information about the R-sig-mixed-models mailing list