gfoRmula: Parametric G-Formula

Implements the parametric g-formula algorithm of Robins (1986) <doi:10.1016/0270-0255(86)90088-6>. The g-formula can be used to estimate the causal effects of hypothetical time-varying treatment interventions on the mean or risk of an outcome from longitudinal data with time-varying confounding. This package allows: 1) binary or continuous/multi-level time-varying treatments; 2) different types of outcomes (survival or continuous/binary end of follow-up); 3) data with competing events or truncation by death and loss to follow-up and other types of censoring events; 4) different options for handling competing events in the case of survival outcomes; 5) a random measurement/visit process; 6) joint interventions on multiple treatments; and 7) general incorporation of a priori knowledge of the data structure.

Getting started

Package details

AuthorVictoria Lin [aut] (V. Lin and S. McGrath made equal contributions), Sean McGrath [aut, cre] (<https://orcid.org/0000-0002-7281-3516>, V. Lin and S. McGrath made equal contributions), Zilu Zhang [aut], Roger W. Logan [aut], Lucia C. Petito [aut], Jessica G. Young [aut] (<https://orcid.org/0000-0002-2758-6932>, M.A. Hernán and J.G. Young made equal contributions), Miguel A. Hernán [aut] (M.A. Hernán and J.G. Young made equal contributions), 2019 The President and Fellows of Harvard College [cph]
MaintainerSean McGrath <sean_mcgrath@g.harvard.edu>
LicenseGPL-3
Version0.3.1
URL https://github.com/CausalInference/gfoRmula https://arxiv.org/abs/1908.07072
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:
install.packages("gfoRmula")

Try the gfoRmula package in your browser

Any scripts or data that you put into this service are public.

gfoRmula documentation built on March 26, 2020, 6:11 p.m.