triplediff: Triple-Difference Estimators

Implements triple-difference (DDD) estimators for both average treatment effects and event-study parameters. Methods include regression adjustment, inverse-probability weighting, and doubly-robust estimators, all of which rely on a conditional DDD parallel-trends assumption and allow covariate adjustment across multiple pre- and post-treatment periods. The methodology is detailed in Ortiz-Villavicencio and Sant'Anna (2025) <doi:10.48550/arXiv.2505.09942>.

Version: 0.1.0
Depends: R (≥ 4.3)
Imports: BMisc (≥ 1.4.6), data.table (≥ 1.15.0), Matrix (≥ 1.6.1), parallel (≥ 1.4.0), parglm (≥ 0.1.7), Rcpp (≥ 1.0.12), speedglm (≥ 0.3-5)
LinkingTo: Rcpp (≥ 1.0.12)
Suggests: testthat (≥ 3.0.0)
Published: 2025-07-23
DOI: 10.32614/CRAN.package.triplediff
Author: Marcelo Ortiz-Villavicencio [aut, cre], Pedro H. C. Sant'Anna [aut]
Maintainer: Marcelo Ortiz-Villavicencio <marcelo.ortiz at emory.edu>
License: MIT + file LICENSE
URL: http://marcelortiz.com/triplediff/
NeedsCompilation: yes
Citation: triplediff citation info
Materials: README, NEWS
CRAN checks: triplediff results

Documentation:

Reference manual: triplediff.html , triplediff.pdf

Downloads:

Package source: triplediff_0.1.0.tar.gz
Windows binaries: r-devel: triplediff_0.1.0.zip, r-release: triplediff_0.1.0.zip, r-oldrel: triplediff_0.1.0.zip
macOS binaries: r-release (arm64): not available, r-oldrel (arm64): not available, r-release (x86_64): triplediff_0.1.0.tgz, r-oldrel (x86_64): triplediff_0.1.0.tgz

Linking:

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