Estimation of the Average Causal Effect with an Instrumental
主 题: Estimation of the Average Causal Effect with an Instrumental
报告人: Eric Tchetgen Tchetgen (Harvard University)
时 间: 2016-12-19 11：00-12：00
地 点: 理科一号楼 1303
Instrumental variables (IVs) are widely used for estimating causal effects in the presence of unmeasured confounding. Under the standard IV model, however, the average treatment effect (ATE) is only partially identifiable. To address this, we propose novel assumptions that allow for identification of the ATE. Our identification assumptions are clearly separated from model assumptions needed for estimation, so that researchers are not required to commit to a specific observed data model in establishing identification. We then construct multiple estimators that are consistent under three different observed data models, and triply robust estimators that are consistent in the union of these observed data models. We pay special attention to the case of binary outcomes, for which we obtain bounded estimators of the ATE that are guaranteed to lie between -1 and 1. Our approaches are illustrated with simulations and a data analysis evaluating the causal effect of education on earnings. About the speaker: Dr. Eric Tchetgen Tchetgen is Professor of Biostatistics and Epidemiologic Methods with joint appointment in the Departments of Biostatistics and Epidemiology at the Harvard T.H. Chan School of Public Health. He received his Ph.D. from Harvard University in 2006 and B.S. from Yale University in 1999. His research has focused on the development of semiparametric theory to provide flexible statistical methods for designing and analyzing both observational and imperfect experimental studies. His work has been influential in the statistical areas of causal inference, selection bias, and missing data, and the study of gene-environment interaction.