Online Program Home
My Program

Abstract Details

Activity Number: 400
Type: Invited
Date/Time: Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #318037
Title: Instrumental Variable Estimation in Observational Studies
Author(s): Miguel Hernan*
Companies: Harvard
Keywords: causal inference ; observational studies ; instrumental variables

Estimating average causal effects from observational data generally requires adjustment for confounding. Instrumental variable (IV) estimation is being increasingly used for confounding adjustment in observational research, with some common proposed instruments being calendar period, genetic traits, and physician's preference. The validity of these IV estimates requires strong conditions that are unlikely to hold, even approximately, in many applications. This talk will discuss some key shortcomings, including some that are rarely discussed, of IV estimation for causal inference from observational research, explore IV approaches that are more robust to violations of assumptions, and describe reporting guidelines for IV analyses.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association