Identification of Risk Groups in Pharmacovigilance Using Penalized Regression and Machine Learning (RGP)

RGP is an R package for analyzing healthcare claims data and simulated data using penalized regression and machine learning methods. This package contains function wrappers to create a simulated cohort, group predictors based on functional targets (from KEGG and TTD) and conventional groups (ATC/ICD systems) and analyze the data using various types of penalized regression (LASSO) and machine learning methods (random forests and block forests).


  1. cohort simulation (R/sim_create_cohort.R)

  2. functional target-based grouping - KEGG (R/ftarget_db_manager.R)

  3. functional target-based grouping - TTD (R/ftarget_db_manager.R)

  4. penalized regression, group-based analysis and results assessment (R/rgp_grpl.R)

  5. classification measures (R/rgp_classification_measures.R)


See the documentation ? and ? for more info.





We gratefully acknowledge the financial support from the innovation fund (“Innovationsfonds”) of the Federal Joint Committee in Germany (grant number: 01VSF16020).


Mariam R. Rizkallah\ Leibniz Institute for Prevention Research & Epidemiology - BIPS GmbH E-mail: rizkallah-issak [at] leibniz-bips [dot] de

bips-hb/rgp documentation built on Feb. 3, 2021, 11:31 a.m.