triplesmatch: Match Triples Consisting of Two Controls and a Treated Unit or Vice Versa

Attain excellent covariate balance by matching two treated units and one control unit or vice versa within strata. Using such triples, as opposed to also allowing pairs of treated and control units, allows easier interpretation of the two possible weights of observations and better insensitivity to unmeasured bias in the test statistic. Using triples instead of matching in a fixed 1:2 or 2:1 ratio allows for the match to be feasible in more situations. The 'rrelaxiv' package, which provides an alternative solver for the underlying network flow problems, carries an academic license and is not available on CRAN, but may be downloaded from 'GitHub' at <>. The 'Gurobi' commercial optimization software is required to use the two functions [infsentrip()] and [triplesIP()]. These functions are not essential to the main purpose of this package. A free academic license can be obtained at <>. The 'gurobi' R package can then be installed following the instructions at <>.

Version: 1.0.0
Depends: rcbalance, rlemon, stats, graphics, MASS, optmatch, utils, rlang
Suggests: rrelaxiv, testthat (≥ 3.0.0), gurobi, sensitivityfull, informedSen, Matrix
Published: 2023-10-23
DOI: 10.32614/CRAN.package.triplesmatch
Author: Katherine Brumberg [aut, cre, cph]
Maintainer: Katherine Brumberg <kbrum at>
License: GPL-3
NeedsCompilation: no
Materials: README NEWS
CRAN checks: triplesmatch results


Reference manual: triplesmatch.pdf


Package source: triplesmatch_1.0.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): triplesmatch_1.0.0.tgz, r-oldrel (arm64): triplesmatch_1.0.0.tgz, r-release (x86_64): triplesmatch_1.0.0.tgz, r-oldrel (x86_64): triplesmatch_1.0.0.tgz


Please use the canonical form to link to this page.