The R Journal: article published in 2017, volume 9:2

openEBGM: An R Implementation of the Gamma-Poisson Shrinker Data Mining Model
Travis Canida and John Ihrie , The R Journal (2017) 9:2, pages 499-519.

Abstract We introduce the R package openEBGM, an implementation of the Gamma-Poisson Shrinker (GPS) model for identifying unexpected counts in large contingency tables using an empirical Bayes approach. The Empirical Bayes Geometric Mean (EBGM) and quantile scores are obtained from the GPS model estimates. openEBGM provides for the evaluation of counts using a number of different methods, including the model-based disproportionality scores, the relative reporting ratio (RR), and the proportional reporting ratio (PRR). Data squashing for computational efficiency and stratification for confounding variable adjustment are included. Application to adverse event detection is discussed.

Received: 2017-08-11; online 2017-11-22, supplementary material, (5.1 KiB)
CRAN packages: openEBGM, PhViD, mederrRank, tidyr, ggplot2, data.table
CRAN Task Views implied by cited CRAN packages: Bayesian, Finance, Graphics, HighPerformanceComputing, Phylogenetics

This article and supplementary materials are licensed under a Creative Commons Attribution 4.0 International license.

@article{RJ-2017-063,
  author = {Travis Canida and John Ihrie},
  title = {{openEBGM: An R Implementation of the Gamma-Poisson Shrinker
          Data Mining Model}},
  year = {2017},
  journal = {{The R Journal}},
  doi = {10.32614/RJ-2017-063},
  url = {https://doi.org/10.32614/RJ-2017-063},
  pages = {499--519},
  volume = {9},
  number = {2}
}