MatchIt: Nonparametric Preprocessing for Parametric Causal Inference
By Daniel E. Ho, Kosuke Imai, Gary King, Elizabeth A. Stuart. 2011
4,858 citations recorded by the source corpus.
MatchIt implements the suggestions of Ho, Imai, King, and Stuart (2007) for improving parametric statistical models by preprocessing data with nonparametric matching methods. MatchIt implements a wide range of sophisticated matching methods, making it possible to greatly reduce the dependence of causal inferences on hard-to-justify, but commonly made, statistical modeling assumptions. The software also easily fits into existing research practices since, after preprocessing data with MatchIt , researchers can use whatever parametric model they would have used without MatchIt , but produce inferences with substantially more robustness and less sensitivity to modeling assumptions. MatchIt is an R program, and also works seamlessly with Zelig .
Subfields: Artificial Intelligence, Statistics and Probability, Causal Inference & Discovery
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
- Cynthia Rudin
- Mohit Bansal
- Alexander Volfovsky
- Dan Jurafsky
- Yoshua Bengio
- Sudeepa Roy
- Noah A. Smith
- Percy Liang
In Statistics and Probability
- Cynthia Rudin
- Alexander Volfovsky
- Peter C. Austin
- Sudeepa Roy
- Jennifer L. Hill
- Elizabeth A. Stuart
- Harsh Parikh
- José R. Zubizarreta
In Causal Inference & Discovery
- Cynthia Rudin
- Alexander Volfovsky
- Sudeepa Roy
- Jennifer L. Hill
- Peter C. Austin
- Elizabeth A. Stuart
- Yameng Liu
- Harsh Parikh
Citations in the corpus
- Racial disparities in automated speech recognition (2020, methodology): “Matching was done with the R package MatchIt (33), with propensity scores estimated via a logistic regression model on the combined data from black and white speakers.”
- MALTS: Matching After Learning to Stretch (2018, background): “Zhao (2004) and Imbens (2004) discuss the choice of distance metric for matching.”
- Practice of Epidemiology Propensity Score Methods for Analyzing Observational Data Like Randomized Experiments: Challenges and Solutions for Rare Outcomes and Exposures (2015, methodology): “All matching was performed using the program MatchIt in R (23, 24).”
- Untangling the Causal Effects of Sex on Judging (2010, methodology): “…that has the closest propensity score is selected.28 We implemented this approach by matching observations from the control group (e.g., male judges on all-male panels) multiple times (“with replacement”).29 28We used the MATCHIT package in R written by Ho et al. (2006) to perform the matching.”
- Prognostic score–based balance measures for propensity score methods in comparative effectiveness research (2013, positive): “1; subclassification and full matching were implemented using the MatchIt and optmatch packages, respectively [18,19].”
- Bayesian Nonparametric Modeling for Causal Inference (2011, methodology): “MatchIt (Ho et al. 2010) in R though MatchIt calls Hansen’s optmatch program for optimal and full matching (Hansen 2004).”
- Space can substitute for time in predicting climate-change effects on biodiversity (2013, methodology): “We processed the Late Quaternary (21–0 kyBP), Late Pleistocene (21–11 kyBP), and Holocene (10–0 kyBP) training and evaluation datasets using coarsened exact matching with the R package MatchIt (53, 54).”
- On the Go: How Mobile Shopping Affects Customer Purchase Behavior (2015, methodology): “We employ 2:1 matching (2 non-M-shoppers to 1 M-shopper) with the nearest-neighbor matching algorithm, which is one of the most widely used methods (Ho et al. 2007; Stuart 2010).”
- Gender differences and bias in open source: pull request acceptance of women versus men (2017, background): “[16] Daniel Ho, Kosuke Imai, Gary King, and Elizabeth Stuart.”