propensity: A Toolkit for Calculating and Working with Propensity Scores

Calculates propensity score weights for multiple causal 'estimands' across binary, continuous, and categorical exposures. Provides methods for handling extreme propensity scores through trimming, truncation, and calibration. Includes inverse probability weighted estimators that correctly account for propensity score estimation uncertainty.

Version:	0.1.0
Depends:	R (≥ 4.2.0)
Imports:	cli, lifecycle, rlang, stats, tidyselect, vctrs (≥ 0.6.5)
Suggests:	dplyr, ggplot2, knitr, mgcv, nnet, parsnip, probably, PSweight, rmarkdown, spelling, testthat (≥ 3.0.0), tibble, tidyr, WeightIt, withr
Published:	2026-03-02
DOI:	10.32614/CRAN.package.propensity (may not be active yet)
Author:	Malcolm Barrett [aut, cre, cph]
Maintainer:	Malcolm Barrett <malcolmbarrett at gmail.com>
BugReports:	https://github.com/r-causal/propensity/issues
License:	MIT + file LICENSE
URL:	https://r-causal.github.io/propensity/, https://github.com/r-causal/propensity
NeedsCompilation:	no
Language:	en-US
Materials:	README
CRAN checks:	propensity results

Documentation:

Reference manual:	propensity.html , propensity.pdf
Vignettes:	Getting Started with propensity (source, R code)

Downloads:

Package source:	propensity_0.1.0.tar.gz
Windows binaries:	r-devel: propensity_0.1.0.zip, r-release: not available, r-oldrel: not available
macOS binaries:	r-release (arm64): propensity_0.1.0.tgz, r-oldrel (arm64): propensity_0.1.0.tgz, r-release (x86_64): propensity_0.1.0.tgz, r-oldrel (x86_64): propensity_0.1.0.tgz

Linking:

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