Robust inference on population indirect causal effects: the generalized front-door criterion

Fulcher, Isabel R.; Shpitser, Ilya; Marealle, Stella; Tchetgen, Eric J. Tchetgen

doi:10.48550/arxiv.1711.03611

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conveniently, most of the estimators for the NIE available in standard software will be robust to unmeasured confounding of this form VanderWeele, 2013, 2015;Tchetgen Tchetgen, 2013;Imai et al, 2010a;Tchetgen Tchetgen and Shpitser, 2012). For investigators who do not wish to make the no interaction assumption, one may consider instead estimating a different form of indirect effect known as the population intervention indirect effect (PIIE) developed by Fulcher et al (2017).…”

Section: Discussionmentioning

confidence: 99%

“…The exclusion restriction likely holds in the context of medication-mediated effects, where the exposure is typically disease status, mediator is medication taken for the disease, and outcome is an unintended effect of medication, so that healthy persons are usually excluded from receiving medication for the disease in question. In a separate strand of work,Fulcher et al (2017) propose a new form of indirect effect, the population intervention indirect effect (PIIE).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Estimation of Natural Indirect Effects Robust to Unmeasured Confounding and Mediator Measurement Error

2019

Self Cite

View full text Add to dashboard Cite

The use of causal mediation analysis to evaluate the pathways by which an exposure affects an outcome is widespread in the social and biomedical sciences. Recent advances in this area have established formal conditions for identification and estimation of natural direct and indirect effects. However, these conditions typically involve stringent no unmeasured confounding assumptions and that the mediator has been measured without error. These assumptions may fail to hold in practice where mediation methods are often applied. The goal of this paper is two-fold. First, we show that the natural indirect effect can in fact be identified in the presence of unmeasured exposure-outcome confounding provided there is no additive interaction between the mediator and unmeasured confounder(s). Second, we introduce a new estimator of the natural indirect effect that is robust to both classical measurement error of the mediator and unmeasured confounding of both exposure-outcome and mediator-outcome relations under certain no interaction assumptions. We provide formal proofs and a simulation study to demonstrate our results. arXiv:1808.03692v1 [stat.ME]

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Estimation of Natural Indirect Effects Robust to Unmeasured Confounding and Mediator Measurement Error

2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…To our knowledge, graphical causal models are currently the only causal inference framework in which it is possible to state a set of causal assumptions and derive from those assumptions a conclusive answer to the question whether, and how, a causal effect of interest is identifiable via covariate adjustment. Covariate adjustment is not complete for identification; other methods like the front-door criterion [6] or docalculus [1] can permit identification even if covariate adjustment is impossible. When multiple options are available, however, adjustment may be an attractive option for effect estimation because its statistical properties are well understood, giving access to useful methodology like robust estimators and confidence intervals.…”

Section: Introductionmentioning

confidence: 99%

Separators and adjustment sets in causal graphs: Complete criteria and an algorithmic framework

Zander

Liśkiewicz

Textor

2019

Artificial Intelligence

View full text Add to dashboard Cite

Principled reasoning about the identifiability of causal effects from non-experimental data is an important application of graphical causal models. This paper focuses on effects that are identifiable by covariate adjustment, a commonly used estimation approach. We present an algorithmic framework for efficiently testing, constructing, and enumerating m-separators in ancestral graphs (AGs), a class of graphical causal models that can represent uncertainty about the presence of latent confounders. Furthermore, we prove a reduction from causal effect identification by covariate adjustment to m-separation in a subgraph for directed acyclic graphs (DAGs) and maximal ancestral graphs (MAGs). Jointly, these results yield constructive criteria that characterize all adjustment sets as well as all minimal and minimum adjustment sets for identification of a desired causal effect with multiple exposures and outcomes in the presence of latent confounding. Our results extend several existing solutions for special cases of these problems. Our efficient algorithms allowed us to empirically quantify the identifiability gap between covariate adjustment and the do-calculus in random DAGs and MAGs, covering a wide range of scenarios. Implementations of our algorithms are provided in the R package DAGITTY.

show abstract

“…In the present work, we provide semiparametric estimators for the average causal effect of a single treatment variable on a single outcome variable in increasingly general scenarios, culminating in semiparametric estimators for any hidden variable causal model of a DAG in which this effect is identifiable. Weight-based estimators for a subclass of models considered in this paper, were studied in (Jung et al, 2020), and to the best of our knowledge, the front-door model (Pearl, 1995) is the only graphical model with unmeasured confounders for which an influence function based estimator has been derived (Fulcher et al, 2017). Other related work includes numerical procedures for approximating the influence function proposed by (Frangakis et al, 2015;Carone et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Semiparametric Inference For Causal Effects In Graphical Models With Hidden Variables

Bhattacharya¹,

Nabi²,

Shpitser³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

The last decade witnessed the development of algorithms that completely solve the identifiability problem for causal effects in hidden variable causal models associated with directed acyclic graphs. However, much of this machinery remains underutilized in practice owing to the complexity of estimating identifying functionals yielded by these algorithms. In this paper, we provide simple graphical criteria and semiparametric estimators that bridge the gap between identification and estimation for causal effects involving a single treatment and a single outcome. First, we provide influence function based doubly robust estimators that cover a significant subset of hidden variable causal models where the effect is identifiable. We further characterize an important subset of this class for which we demonstrate how to derive the estimator with the lowest asymptotic variance, i.e., one that achieves the semiparametric efficiency bound. Finally, we provide semiparametric estimators for any single treatment causal effect parameter identified via the aforementioned algorithms. The resulting estimators resemble influence function based estimators that are sequentially reweighted, and exhibit a partial double robustness property, provided the parts of the likelihood corresponding to a set of weight models are correctly specified. Our methods are easy to implement and we demonstrate their utility through simulations.

show abstract

Robust inference on population indirect causal effects: the generalized front-door criterion

Cited by 3 publications

References 0 publications

Estimation of Natural Indirect Effects Robust to Unmeasured Confounding and Mediator Measurement Error

Estimation of Natural Indirect Effects Robust to Unmeasured Confounding and Mediator Measurement Error

Separators and adjustment sets in causal graphs: Complete criteria and an algorithmic framework

Semiparametric Inference For Causal Effects In Graphical Models With Hidden Variables

Contact Info

Product

Resources

About