Article ID Journal Published Year Pages File Type
7547724 Statistical Methodology 2014 4 Pages PDF
Abstract
We show that assumptions that are sufficient for estimating an average treatment effect in randomized trials with non-compliance restrict the subgroup means for always takers, compliers, defiers and never takers to a two-dimensional linear subspace of a four-dimensional space. Implications and special cases are exemplified.
Related Topics
Physical Sciences and Engineering Mathematics Statistics and Probability
Authors
,