biometrician / abe-for-sas

Augmented backward elimination for SAS: A Pragmatic and Purposeful Way to Develop Statistical Models

abe_for_sas

SAS-Macros Implementing Augmented Backward Elimination: A Pragmatic and Purposeful Way to Develop Statistical Models

Description

Here we provide our SAS-macros to apply augmented backward elimination in linear, logistic or Cox-proportional hazards regression models.

• ABE.SAS: The macro %abe implements the augmented backward elimination algorithm.
• ABE_BOOTSTRAP.SAS: The macro %abe_bootstrap implements abe analysis for bootstrap samples.

If you use the macro in a scientific report, please cite our corresponding paper (Dunkler, Plischke, Leffondre and Heinze, 2014). In the supplementary information of the paper, the user will find a technical report with a detailed explanation of the parameters of the SAS macros and a worked example (code and data).

Acknowledgment

This work was supported by the European Community's Seventh Framework Programme, grant agreement number 241544.

References

Dunkler D, Plischke M, Leffondre K, Heinze G: Augmented Backward Elimination: A Pragmatic and Purposeful Way to Develop Statistical Models. PLoS ONE (2014) 9(11):e113677 DOI:10.1371/journal.pone.0113677.