Efficient estimation of average treatment effects using the estimated propensity score

By Keisuke Hirano, Guido W. Imbens, Geert Ridder. 2003

2,814 citations recorded by the source corpus.

Subfields: Economics and Econometrics, 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 Economics and Econometrics

In Statistics and Probability

In Causal Inference & Discovery

Citations in the corpus

Recommendations as Treatments: Debiasing Learning and Evaluation (2016, background): “This effect is known, partly because the estimated propensities can provide an effect akin to stratification (Hirano et al., 2003; Wooldridge, 2007).”
Approximate residual balancing: debiased inference of average treatment effects in high dimensions (2016, methodology): “Such inverse-propensity weights with a flexibly estimated propensity score have desirable asymptotic properties [Hirano et al., 2003] in settings where the asymptotics is based on a fixed number of covariates.”
Doubly Robust Off-policy Value Evaluation for Reinforcement Learning (2015, background): “In medical and public-policy domains (Murphy et al., 2001; Hirano et al., 2003), it is often hard to run a controlled experiment to estimate the treatment effect, and off-policy There are roughly two classes of approaches to off-policy value evaluation.”
Estimation and Inference of Heterogeneous Treatment Effects using Random Forests (2015, methodology): “Thus, if we knew e(x), we would have access to a simple unbiased estimator for τ (x); this observation lies at the heart of methods based on propensity weighting (e.g., Hirano, Imbens, and Ridder 2003).”
Efficiently Breaking the Curse of Horizon in Off-Policy Evaluation with Double Reinforcement Learning (2019, background): “In ﬁnite-horizon problems ( i.e. , when the estimand is ρ π e T ), when the behavior policy is known, IS is unbiased and consistent but its variance tends to be large and it is ineﬃcient (Hirano et al., 2003).”
Double Reinforcement Learning for Efficient Off-Policy Evaluation in Markov Decision Processes (2019, background): “In particular, under M1, IS with λ̂t = λt is known to be inefficient (Hirano et al., 2003), which implies it must be inefficient under M2 as well.”
Towards Optimal Off-Policy Evaluation for Reinforcement Learning with Marginalized Importance Sampling (2019, methodology): “IS-based off-policy evaluation methods have also seen lots of interest recently especially for short-horizon problems, including contextual bandits [Murphy et al., 2001; Hirano et al., 2003; Dudík et al., 2011; Wang et al., 2017].”
Covariate balancing propensity score for a continuous treatment: Application to the efficacy of political advertisements (2018, positive): “1 Introduction Propensity score methods are popular among researchers who wish to infer causal effects in observational studies (e.g., Rosenbaum and Rubin, 1983, 1984, 1985; Robins et al., 2000; Hirano et al., 2003).”
Marks of Distinction (2015, methodology): “The method is an extension of inverse probability treatment weights (IPTW; see, for example, Hirano, Imbens, and Ridder, 2003), whereby the odds ratios of the estimated propensity score are used as weights for the control-group observations (non-ranked) and the weights of observations in the…”
Irregular Identification, Support Conditions, and Inverse Weight Estimation (2010, background): “(ii) 0 < P (di = 1|xi) < 1 ∀xi See also Hirano, Imbens, and Ridder (2003). The above assumption can be used to identify β as β = EX [E[yi|di = 1, xi]− E[yi|di = 0, xi]] (4.”