Double Robustness in Estimation of Causal Treatment Effects
Double Robustness in Estimation of Causal Treatment Effects
Double Robustness in Estimation of Causal Treatment Effects
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
2. Counterfactual Model<br />
Unconditional (marg<strong>in</strong>al) expectation: Conceptually, the “average ”<br />
across all possible values a random variable can take on <strong>in</strong> the population<br />
Statistical <strong>in</strong>dependence: For two random variables Y and Z<br />
• Y and Z are <strong>in</strong>dependent if the probabilities with which Y takes on<br />
its values are the same regardless <strong>of</strong> the value Z takes on<br />
• Notation – Y � Z<br />
<strong>Double</strong> <strong>Robustness</strong>, EPID 369, Spr<strong>in</strong>g 2007 8