approxmatch-package {approxmatch} | R Documentation |
Approximately Optimal Fine Balance Matching with Multiple Groups.
Description
Tools for constructing a matched design with multiple comparison groups. Further specifications of refined covariate balance restriction and exact match on covariate can be imposed. Matches are approximately optimal in the sense that the cost of the solution is at most twice the optimal cost, Crama and Spieksma (1992) <doi:10.1016/0377-2217(92)90078-N>, Karmakar, Small and Rosenbaum (2019) <doi:10.1080/10618600.2019.1584900>.
Details
Index of help topics:
Dodgeram Dodge ram pk 2500 data on side airbag (SAB) usage from 1995 to 2015 approxmatch-package Approximately Optimal Fine Balance Matching with Multiple Groups. covbalance Check covariate balance of a design. kwaymatching Create approximately optimal matched strata of multiple, at least two, groups. multigrp_dist_struc Construct the distance structure for the multiple groups. nrbalancematch The is the background function to perform matching using network optimization.
An R package for creating matched strata with multiple treatments. Default design for a stratum structure is one unit from each treatment, but, other designs can be specified. User can also fine match/ near fine match on one or more categorical covariates, e.g. sex and age group.
The main functions of the package are kwaymatching
and tripletmatch
.
These functions take as input the distance structure of multiple groups and the grouping
information to create an approximately optimal multigroup design minimizing the
total distance. A distance structure can be calculated as per requirement by
the multigrp_dist_struc
function.
The algorithm used to create matched design is an approximation algorithm developed by Karmakar, Small and Rosenbaum (2019). The design built is guaranteed to be close to the optimal matched design of the specified structure.
IMPORTANT NOTE: In order to perform matching, kwaymatching
requires the
user to load the optmatch (>= 0.9-1) package separately. A manual loading is
required due to software license issues. If the package is not loaded, the
kwaymatching
command will fail with an error saying the optmatch package
is not present. Reference to optmatch is given below.
Author(s)
Bikram Karmakar
Maintainer: Bikram Karmakar <bkarmakar@ufl.edu>
References
Crama, Y. and Spieksma, F. C. R. (1992), Approximation algorithms for three-dimensional assignment problems with triangle inequalities, European Journal of Operational Research 60, 273–279.
Hansen, B.B. and Klopfer, S.O. (2006) Optimal full matching and related designs via network flows, JCGS 15 609–627.
Karmakar, B., Small, D. S. and Rosenbaum, P. R. (2019) Using Approximation Algorithms to Build Evidence Factors and Related Designs for Observational Studies, Journal of Computational and Graphical Statistics, 28, 698–709.
Examples
## See kwaymatching for usage