【统计学论坛】Sharp sensitivity bounds for mediation under unmeasured mediator-outcome confounding

Title (题目): Sharp sensitivity bounds for mediation under unmeasured mediator-outcome confounding
Time (时间): 4pm-5pm, 2015-12-03 (Thursday)
Location (地点): 伟清楼209 (Center for Statistical Science, Tsinghua University)
Speaker (报告人): Peng Ding, UC Berkeley

Abstract (摘要):
It is often of interest to decompose a total effect of an exposure into the component that acts on the outcome through some mediator and the component that is independent of it, i.e., through other pathways. Said another way, we are interested in the direct and indirect effects of the exposure on the outcome. However, even if the exposure is randomly assigned, it is often infeasible to randomize the mediator, leaving the mediator-outcome confounding not fully controlled. We develop a sensitivity analysis technique that allows for obtaining bounds on the direct and indirect effects without making parametric assumptions about the unmeasured mediator-outcome confounding.

About the speaker (报告人介绍)
Dr. Ding received his Bachelor’s and Master’s degrees from Peking University. He later received his Master’s and Doctor’s degrees from Harvard University. Starting 2016, He will join the University of California, Berkeley, as an Assistant Professor of Statistics.