A repeatedly measured outcome for subjects in an observational study allows researchers to monitor how the outcome changes over time. When an intervention that affects the outcome is initiated at different time following a guideline, it is essential to account for the varying time to intervention (TTI) in modeling changes in the outcome over time and sample selection bias due to the guideline-based intervention. In this talk, I introduce a TTI-varying coefficient model that describes the population mean outcome trajectory and a double-weighted estimation procedure that corrects estimation bias of the TTI-varying coefficients.
Speaker: Hyunkeun Ryan Cho, PhD, Associate Professor of Biostatistics, University of Iowa
Register: http://tiny.cc/EpiBioEmailList