Causal inference in observational studies with clustered data
Meng Wu · 2016
In this thesis, we study causal inference in observational studies with clustered data.We present the study in three papers in the order of the problems that we propose to solve.In the first paper, we study causal inference using potentially observable framework in clustered data (e.g.intervention studies on patients nested within hospitals).We employ mixed-effects models and sandwich estimator to derive inference on the average causal effect (ACE).Our methods apply the concept of potential outcomes in Rubin Causal Model (RCM) and extend Schafer's method of estimating the variance of ACE.Particularly, we develop two model-based approaches to estimate the ACE under dual-modeling strategy which adjusts for the confounding effect by inverse probability weighting (IPW).These two approaches use the same random intercept linear model for the estimation of potential outcomes, but differ in the treatment assignment model.