On Partial Identification of the Pure Direct Effect
By Caleb H. Miles, Phyllis Kanki, Seema Meloni, E. T. Tchetgen Tchetgen. 2015
15 citations recorded by the source corpus.
In causal mediation analysis, nonparametric identification of the pure (natural) direct effect typically relies on, in addition to no unobserved pre-exposure confounding, fundamental assumptions of (i) so-called "cross-world-counterfactuals" independence and (ii) no exposure- induced confounding. When the mediator is binary, bounds for partial identification have been given when neither assumption is made, or alternatively when assuming only (ii). We extend existing bounds to the case of a polytomous mediator, and provide bounds for the case assuming only (i). We apply these bounds to data from the Harvard PEPFAR program in Nigeria, where we evaluate the extent to which the effects of antiretroviral therapy on virological failure are mediated by a patient's adherence, and show that inference on this effect is somewhat sensitive to model assumptions.
Subfields: 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 Statistics and Probability
- Razieh Nabi
- Iv'an D'iaz
- Nima S. Hejazi
- Kara E. Rudolph
- Ilya Shpitser
- M. J. van der Laan
- Edward H. Kennedy
- Judea Pearl
Citations in the corpus
- Fair Inference on Outcomes (2017, background): “While finding such bounds is an open problem in general, they were derived in the context of the NDE with a discrete mediator in (Miles et al. 2016).”
- Causal mediation analysis for stochastic interventions (2019, methodology): “Point identification of natural (in)direct effects in the presence of such variables is not generally possible, and its partial identification is an area of active research (Robins and Richardson, 2010; Tchetgen Tchetgen and Phiri, 2014; Miles et al., 2015).”
- Nonparametric efficient causal mediation with intermediate confounders (2019, background): “The first class attempts to identify and estimate bounds on the natural (in)direct effects without cross-world counterfactual independencies or in the presence of intermediate confounders (e.g., Robins and Richardson, 2010; Tchetgen and Phiri, 2014; Miles et al., 2015), whereas the second class of m”