Matching on the Estimated Propensity Score

By Alberto Abadie, Guido W. Imbens. 2016

960 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Education, Statistics and Probability

How researchers see this paper

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

In Economics and Econometrics

In Education

In Statistics and Probability

Citations in the corpus

Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations (2020, neutral): “Several ways of handling the selection bias involve using propensity matching (Austin, 2011; Imai & Ratkovic, 2014; Abadie & Imbens, 2016), building representations where treated and un-treated populations had similar distributions (Johansson et al., 2016; Shalit et al., 2017; Li & Fu, 2017; Yao et…”
Deep Counterfactual Networks with Propensity-Dropout (2017, background): “…of inferring individualized treatment effects from observational datasets is a fundamental problem in many domains such as precision medicine (Shalit et al., 2017), econometrics (Abadie & Imbens, 2016), social sciences (Athey & Imbens, 2016), and computational advertising (Bottou et al., 2013).”
Bayesian Nonparametric Causal Inference: Information Rates and Learning Algorithms (2017, methodology): “This dependence is quantified via the conditional distribution P(ωi = 1|Xi = x), also known as the propensity score of subject i [13], [11].”
Asymptotic Variance of Semiparametric Estimators With Generated Regressors (2013, background): “Comparing the correction terms (1) and (2) leads us to an interesting conclusion: The influence function for β is equal to that of the unfeasible estimator β that ignores the estimation error in the first step, i.”
Generalizing from unrepresentative experiments: a stratified propensity score approach (2014, background): “There has been some work on the effect of using estimated propensity scores (see, for example, Abadie and Imbens (2009)).”
A Martingale Representation for Matching Estimators (2012, methodology): “Therefore, the size of the matching discrepancies should be conscientiously described (e.g., using the techniques in Abadie and Imbens 2011) before the result in Theorem 1 is applied.”
Matching for Balance (2019, background): “In another variant, Abadie and Imbens (2016) formalized matching on the estimated propensity score.”
Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes (2017, background): “on Wi ⊥⊥ Xi is satisﬁed only in the special case when D comprises data from a randomized control trial.) The conditional distribution P(Wi = 1|Xi = x), also known as the propensity score of subject i [1, 7, 21], reﬂects the underlying (unknown) policy for assigning the treatment to the subjects. Thr”
Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference (2020, background): “…in practice, the true interference is unknown and we could only match on estimated values of f ; this suﬀers from all the problems that aﬄict matching on estimated propensity scores without appropriate adjustments (Abadie and Imbens, 2016) or parametric approximations (Rubin and Thomas, 1996).”
Matching methods for causal inference: A review and a look forward (2010, background): “There are also some empirical formulas for variance estimation for particular matching scenarios (e.g., Abadie and Imbens, 2006, 2009b; Schafer and Kang, 2008), but this is an area for future research.”
Late-career entrepreneurship, income and quality of life (2017, methodology): “The standard errors of the ATT (in parentheses) are based on Abadie and Imbens (2009) and account for the propensity score being estimated rather than observed. 11 T. Kautonen et al. / Journal of Business Venturing xxx (2017) xxx–xxx”