Concerning the consistency assumption in causal inference.

By Tyler J. VanderWeele. Epidemiology, 2009

368 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/fd8861991e13dcba3e0ce4e57c529a1c9ce4809e

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

Citations in the corpus

A Framework for Dynamic Causal Inference in Political Science (2013, background): “Recent work has further formalized and clarified this assumption (Cole and Frangakis 2009; Pearl 2010; VanderWeele 2009). suppressed.”
Semiparametric theory and empirical processes in causal inference (2015, background): “le, suppose that in a population of interest we actually get to observe potential outcomes under the received treatment for each unit, i.e., A =a =⇒ Y =Ya. (C1) Condition (C1) is called “consistency” [68] and holds if pot ential outcomes are deﬁned uniquely by a unit’s own treatment and not others’ ”
Conceptual issues concerning mediation, interventions and composition (2009, methodology): “…mediator-outcome confounders but estimation in such cases requires techniques beyond simple regression analysis (Robins, 1999; van der Laan and Petersen, 2008; Conceptual issues concerning mediation, interventions and composition 461 Goetgeluk et al., 2008; VanderWeele, 2009a; Vansteelandt, 2009).”
Causal Inference Under Multiple Versions of Treatment (2013, positive): “VanderWeele (2009) referred to [1] as an assumption of treatment variation irrelevance (which need not necessarily imply consistency assumption (2)).”
Causal diagrams for interference (2014, background): “Generalizations of the contagion and infectiousness effects to blocks of size greater than two are also possible (VanderWeele et al., 2012). An alternative definition of an infectiousness effect, proposed by VanderWeele et al. (2012), is the controlled direct effect of A1 on Y T 2 , holding Y T0 1 f”
Geographic Boundaries as Regression Discontinuities (2015, background): “When analysts face compound treatments, an additional assumption will be needed for identification (see VanderWeele 2009; Hernán and VanderWeele 2011).6 We can state this assumption formally as follows.”
Estimating Average Causal Effects Under Interference Between Units (2013, methodology): “We assume the following consistency condition that relates the observed data to potential outcomes under the exposure model (VanderWeele, 2009):”
All your data are always missing: incorporating bias due to measurement error into the potential outcomes framework. (2015, background): “Like the assumption of no measurement error, we generally assume treatment version irrelevance prior to linking the observed outcomes to the potential outcomes by consistency.(11,12,13) In addition, we assumed that each participant’s potential outcomes did not depend on the exposures of other partic”
Generalizing causal inferences from randomized trials: counterfactual and graphical identification (2019, background): “It entails an assumption of no hidden versions of the intervention [28, 29] or “treatment variation irrelevance” [30], that is to say, either there is only one way to intervene on R, S, and Z, or the different ways of intervening have the same effect on the outcome.”