causalOT: Optimal Transport Weights for Causal Inference

Uses optimal transport distances to find probabilistic matching estimators for causal inference. These methods are described in Dunipace, Eric (2021) <arXiv:2109.01991>. The package will build the weights, estimate treatment effects, and calculate confidence intervals via the methods described in the paper. The package also supports several other methods as described in the help files.

Version: 0.1.2
Depends: R (≥ 3.5.0)
Imports: approxOT, Matrix, matrixStats, methods, lbfgsb3c, loo, osqp, pbapply, reticulate, R6 (≥ 2.4.1), Rcpp (≥ 1.0.3), RSpectra, sandwich
LinkingTo: BH (≥ 1.66.0), Rcpp (≥ 0.12.0), RcppEigen (≥ 0.3.3.3.0)
Suggests: CBPS, data.table (≥ 1.12.8), rstan (≥ 2.19.3), Rmosek, testthat (≥ 2.1.0), knitr, rmarkdown
Published: 2022-09-04
Author: Eric Dunipace ORCID iD [aut, cre]
Maintainer: Eric Dunipace <edunipace at mail.harvard.edu>
License: GPL (≥ 3.0)
NeedsCompilation: yes
Citation: causalOT citation info
Materials: README
CRAN checks: causalOT results

Documentation:

Reference manual: causalOT.pdf
Vignettes: Using causalOT

Downloads:

Package source: causalOT_0.1.2.tar.gz
Windows binaries: r-devel: causalOT_0.1.2.zip, r-release: causalOT_0.1.2.zip, r-oldrel: causalOT_0.1.2.zip
macOS binaries: r-release (arm64): causalOT_0.1.tgz, r-oldrel (arm64): causalOT_0.1.tgz, r-release (x86_64): causalOT_0.1.2.tgz, r-oldrel (x86_64): causalOT_0.1.2.tgz
Old sources: causalOT archive

Linking:

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