Multivariate and Propensity Score Matching Software with Automated Balance Optimization: The<b>Matching</b>Package for<i>R</i>

TLDR

Matching is an R package that provides multivariate and propensity score matching functions and optimizes covariate balance using a genetic search algorithm, offering a range of univariate and multivariate metrics to assess balance. Its core algorithm, written in C++ and leveraging BLAS, scales efficiently with dataset size, while a parallelized genetic algorithm enables optimal balance across multiple CPUs or clusters, and users can fine‑tune matching and balance evaluation through numerous options.

Abstract

<b>Matching</b> is an R package which provides functions for multivariate and propensity score matching and for finding optimal covariate balance based on a genetic search algorithm. A variety of univariate and multivariate metrics to determine if balance actually has been obtained are provided. The underlying matching algorithm is written in C++, makes extensive use of system BLAS and scales efficiently with dataset size. The genetic algorithm which finds optimal balance is parallelized and can make use of multiple CPUs or a cluster of computers. A large number of options are provided which control exactly how the matching is conducted and how balance is evaluated.