The MIT Sociotechnical Systems Research Center invites you to its joint seminar with the Laboratory for Information and Decision Systems on Tuesday, March 3, at 4 p.m. in Room 32-155. Professor Alberto Abadie from Harvard University will present "Endogenous Stratification in Randomized Experiments." A reception will follow.
Researchers and policy makers are often interested in estimating how policy interventions affect the outcomes of those individuals most in need of help. This concern has motivated the widespread practice of estimating policy effects separately by brackets of the predicted outcomes that the individuals in a sample would attain in the absence of a policy intervention. Abadie will explain how substantial biases may arise in practice if predicted outcomes in the absence of the intervention are estimated, as is often the case, using the full sample of individuals not exposed to the intervention. He will discuss the behavior of leave-one-out and repeated split sample estimators and show that they behave well in realistic scenarios, correcting the large bias problem of the full sample estimator. He will present data from the National JTPA Study and the Tennessee STAR experiment to demonstrate the performance of alternative estimators and the magnitude of their biases.