Identifiability of Path-Specific Effects

By Chen Avin, Ilya Shpitser, Judea Pearl. 2005

381 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 Causal Inference & Discovery

Citations in the corpus

Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias (2020, background): “The causality literature often focuses on assumptions needed for identification of mediation effects from observed data (Pearl, 2001; Avin et al., 2005; Imai et al., 2010b).”
Causal Modeling-Based Discrimination Discovery and Removal: Criteria, Bounds, and Algorithms (2019, methodology): “To this end, we employ the technique of the path-specific effect [11], [12].”
A Causal Framework for Discovering and Removing Direct and Indirect Discrimination (2016, methodology): “To measure the causal e ﬀ ect transmitted along a certain causal path, we employ the formulation of the path-speciﬁc e ﬀ ect (Avin, Shpitser, and Pearl 2005; Shpitser 2013).”
Causal inference in statistics: An overview (2009, background): “Such applications motivate the analysis of path-specific effects, that is, the effect of X on Y through a selected set of paths (Avin et al., 2005).”
A general approach to causal mediation analysis. (2010, neutral): “Baron and Kenny (1986) did not make this assumption, and allowed the variable acting on the mediator and (in the case of partial mediation) outcome variable to be continuous.”
Quantifying an Adherence Path-Specific Effect of Antiretroviral Therapy in the Nigeria PEPFAR Program (2014, background): “Effects along such arbitrary (bundles of) causal pathways are known as path-specific effects (Pearl, 2000; Avin et al., 2005; Shpitser, 2013) and it is possible to define them inductively, which results in a quantity that is a function of a nested counterfactual (Shpitser, 2013).”
Interpretation and identification of causal mediation. (2014, positive): “For nonparametric models, Avin et al. (2005) derived a necessary and sufficient condition for identifying (natural) path-specific effects in any graph structure with no unmeasured confounders.”
IDENTIFYING CAUSAL MECHANISMS (PRIMARILY) BASED ON INVERSE PROBABILITY WEIGHTING (2014, background): “If the latter is the case, the identification of the ‘total’ indirect effect, which also accounts for correlations between those covariates affected by the treatment and the mediator, requires additional restrictions (see Robins, 2003; Avin et al., 2005; and Imai and Yamamoto, 2013).”
On Partial Identification of the Pure Direct Effect (2015, background): “The study of causal mediation has in recent years enjoyed an explosion in popularity (Robins and Greenland, 1992; Robins, 1999, 2003; Pearl, 2001; Avin et al., 2005; Taylor et al., 2005; Petersen et al., 2006; Ten Have et al., 2007; Albert, 2008; Goetgeluk et al., 2008; van der Laan and Petersen, 20”
The Causal Foundations of Structural Equation Modeling (2012, background): “Such applications motivate the analysis of path-specific effects—that is, the effect of X on Y through a selected set of paths (Avin et al., 2005). In all these cases, the policy intervention invokes the selection of signals to be sensed rather than variables to be fixed. Pearl (2001) has therefore ”
Estimating Causal Effects by Bounding Confounding (2014, methodology): “Note that in mediation analysis [Pearl, 2001, Avin et al., 2005, Robins and Greenland, 1992] a similar splitting into direct and indirect effect is used.”
Conceptual issues concerning mediation, interventions and composition (2009, background): “However, it has been shown that whenever there is a consequence of the exposure that also affects both the mediator and the outcome then natural direct and indirect effects are not in general identified (Avin et al., 2005).”
Causal diagrams for interference (2014, background): “Avin et al. (2005) proved that in most settings the following is a necessary assumption for the identification of the natural direct and indirect effects: there is no variable Wi such that (i) there is an activated directed path from Ai to Wi, (ii) there is a deactivated directed path from Wi to Yi,”
Odds ratios for mediation analysis for a dichotomous outcome. (2010, background): “If, however, there is a variable L that is an effect of A and affects both M and Y, then assumption 4 is violated and natural direct and indirect effects will not in general be identified (17), irrespective of whether data are available on L.”
Nonparametric efficient causal mediation with intermediate confounders (2019, background): “Unfortunately, natural direct and indirect effects are not generally identified in the presence of a mediator-outcome confounder affected by treatment such as in the DAG in Figure (1) (Avin et al., 2005; Tchetgen and VanderWeele, 2014).”
Causal mediation analysis for stochastic interventions (2019, methodology): “A comprehensive discussion of issues in identification of path effects that includes this issue as a particular problem may be found in Avin et al. (2005). VanderWeele et al.”