On model selection and model misspecification in causal inference

185 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.

Estimation of direct effects for survival data by using the Aalen additive hazards model (2011, background): “This is because the resulting bias is a form of ‘M-bias’, which has been found to be typically small in other contexts (Greenland, 2003; Vansteelandt et al., 2010).”
Globally efficient non-parametric inference of average treatment effects by empirical balancing calibration weighting (2015, methodology): “…have proposed estimators by combining equations (1) and (4) in various creative manners under the propensity score framework; see Qin and Zhang (2007), Tan (2010), Chan (2012), Graham et al. (2012), Vansteelandt et al. (2012), Han and Wang (2013), Imai and Ratkovic (2014) and Chan and Yam (2014).”
Bias-Reduced Doubly Robust Estimation (2015, methodology): “This is motivated by the fact that the bias of a DR estimator can get severely amplified under misspecification of at least one working model and may become especially severe under misspecification of both working models (Kang and Schafer 2007a; Vansteelandt, Bekaert, and Claeskens 2012).”