Causal Inference With General Treatment Regimes

By Kosuke Imai, David A. van Dyk. 2004

1,011 citations recorded by the source corpus.

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders (2019, positive): “Assumption 2: Positivity (Overlap) (Imai & Van Dyk, 2004): If P (Āt−1 = āt−1, X̄t = x̄t) 6= 0 then P (Āt = āt | Āt−1 = āt−1, X̄t = x̄t) > 0 for all āt.”
The State of Applied Econometrics - Causality and Policy Evaluation (2016, neutral): “Imbens [2003] builds on the Rosenbaum and Rubin approach by developing a data-driven way to obtain a set of correlations between the unobserved covariates and treatment and outcome.”
Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations (2020, background): “If A t = a t for a given patient, then the potential outcome for treatment a t is the same as the observed (factual) outcome: Assumption 2: Positivity (Overlap) (Imai & Van Dyk, 2004): Assumption 2 means that, for each timestep, each treatment has non-zero probability of being assigned.”
Can Nonrandomized Experiments Yield Accurate Answers? A Randomized Experiment Comparing Random and Nonrandom Assignments (2008, methodology): “One useful and increasingly common approach is to construct a simulation that uses data observed in an actual study (see, e.g., Imai and van Dyk 2004; Hill and McCulloch 2008).”
The Dangers of Extreme Counterfactuals (2006, background): “…medical researchers are almost always interested in h, political scientists are also interested in the average causal effect for the entire set of observations, c5EðY1Þ EðY0Þ; ð12Þ 13The analysis of treatments with more than two levels follows analogously (e.g., Imai and Dyk 2004; Lechner 1999).”
Uniformly Semiparametric Efficient Estimation of Treatment Effects With a Continuous Treatment (2015, methodology): “Hirano, Imbens, and Ridder (2003) assume the second moment of Y (1) and Y (0) to be finite, which is slightly stronger than our condition.”
Too much of a good thing? On the growth effects of the EU's regional policy (2012, background): “The binary case was extended to categorial multivalued treatment by Imbens (2000) and, more recently, to continuous treatments (see Hirano and Imbens, 2004; Imai and Van Dijk, 2004).”
Randomization Inference With Natural Experiments (2006, background): “…of estimating the effect of being listed on the first page, it may also be possible to analyze the effect of being listed on each page as a multitreatment regime by assuming a constant additive treatment effect (see, e.g., Angrist and Imbens 1995; Imai and van Dyk 2004; Imbens and Rosenbaum 2005).”
Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference (2007, methodology): “In simulations, the reduction in n required to overcome the advantages in mean squared error is usually quite substantial (Rubin and Thomas 1996; Imai and van Dyk 2004).”