riskCommunicator: G-Computation to Estimate Interpretable Epidemiological Effects

Estimates flexible epidemiological effect measures including both differences and ratios using the parametric G-formula developed as an alternative to inverse probability weighting. It is useful for estimating the impact of interventions in the presence of treatment-confounder-feedback. G-computation was originally described by Robbins (1986) <doi:10.1016/0270-0255(86)90088-6> and has been described in detail by Ahern, Hubbard, and Galea (2009) <doi:10.1093/aje/kwp015>; Snowden, Rose, and Mortimer (2011) <doi:10.1093/aje/kwq472>; and Westreich et al. (2012) <doi:10.1002/sim.5316>.

Version: 0.1.0
Depends: R (≥ 3.5)
Imports: boot, dplyr, ggplot2, gridExtra, magrittr, purrr, stats, rlang, tidyr, tidyselect, tidyverse
Suggests: knitr, rmarkdown, testthat, printr
Published: 2020-06-26
Author: Jessica Grembi ORCID iD [aut, cre, cph], Elizabeth Rogawski McQuade ORCID iD [ctb]
Maintainer: Jessica Grembi <jess.grembi at gmail.com>
License: GPL-3
NeedsCompilation: no
Materials: README
CRAN checks: riskCommunicator results

Downloads:

Reference manual: riskCommunicator.pdf
Vignettes: riskCommunicator package vignette for manuscript
riskCommunicator package extended vignette for newbie R users
Package source: riskCommunicator_0.1.0.tar.gz
Windows binaries: r-devel: riskCommunicator_0.1.0.zip, r-release: riskCommunicator_0.1.0.zip, r-oldrel: riskCommunicator_0.1.0.zip
macOS binaries: r-release: riskCommunicator_0.1.0.tgz, r-oldrel: riskCommunicator_0.1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=riskCommunicator to link to this page.