Multivariate and Propensity Score Matching Software with Automated Balance Optimization: The Matching Package for R

By Jasjeet S. Sekhon. 2008

905 citations recorded by the source corpus.

Matching 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.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/9e90da842b42e5dd678dce65246683620cd71dc9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Automated versus Do-It-Yourself Methods for Causal Inference: Lessons Learned from a Data Analysis Competition (2017, neutral): “A strong focus of the causal inference literature over the past few decades has been on preprocessing data to reduce reliance on parametric assumptions (Scharfstein, Rotnitzky and Robins, 1999; Bang and Robins, 2005; Sekhon, 2007).”
The International Journal of Biostatistics (2017, critical): “The popular propensity score methods to assess causal eﬀects of single time point interventions (e.g., Rosenbaum and Rubin (1983), Sekhon (2008), Rubin (2006)) are not double robust (i.e., rely on correct spec-iﬁcation of propensity score), have no natural generalization to multiple time-point…”
Classical boson sampling algorithms with superior performance to near-term experiments (2017, neutral): “More rigorously, we performed 2-sample bootstrap Kolmogorov-Smirnov (KS) tests [27, 28] comparing the distribution of − log(|Per(AS)|(2)) values for tuples obtained by sampling from the distinguishable particle distribution with those obtained using our classical boson sampling methods.”
A Primer on Inverse Probability of Treatment Weighting and Marginal Structural Models (2016, neutral): “Some researchers (van der Laan, Polley, & Hubbard, 2007) strongly prefer data mining techniques to break the reliance on parametric models (for a review, see McCaffrey, Ridgeway, & Morral, 2004; Lee, Lessler, & Stuart, 2010; Sekhon, 2011; Imai & Ratkovic, 2014; or Westreich, Lessler, & Jonsson Funk,”
Newspaper Closures Polarize Voting Behavior (2018, positive): “NEWSPAPER CLOSURES 33 Tables NEWSPAPER CLOSURES 34 NEWSPAPER CLOSURES 35 Supplementary Material “Newspaper Closures Polarize Voting Behavior” Matching Balance Statistics: Placebo Test We matched on the same covariates as in the analysis in the main text, exact matching on states and utilizing the ge”