Why prefer double robust estimators in causal inference

132 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.

An application of model-fitting procedures for marginal structural models. (2005, background): “Treatment weights defined as the inverse of the conditional probability give the most weight to the observations that are the most unusual (where ‘‘usual’’ is defined by the ideal experiment in which A is randomized), hence the term ‘‘inverse probability of treatment weights’’ ( 5 ).”
A Review of Generalizability and Transportability (2021, methodology): “…data) but should be driven by similar principles as those that guide the choice between outcome regressions, matching and weighting methods, and double robust approaches for confounding adjustment (Van der Laan, Laan and Robins, 2003; Neugebauer and van der Laan, 2005; van der Laan and Rose, 2011).”
A Tutorial on Propensity Score Estimation for Multiple Treatments Using Generalized Boosted Models (2013, background): “also be more efficient than the simple weighted estimator [39].”