JSM 2012 Home

JSM 2012 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Online Program Home

Abstract Details

Activity Number: 304
Type: Contributed
Date/Time: Tuesday, July 31, 2012 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics in Epidemiology
Abstract - #305391
Title: An Unidentifiability Issue in G-Estimation and Suggestions to Fix It
Author(s): Yang Jiang*+
Companies: The Wharton School
Address: 329 S. 42nd Street, Philadelphia, PA, 19104, United States
Keywords: g-estimation ; counterfactual outcome ; unidentifiability ; future ignorability assumption

In causal inference for longitudinal data, g-estimation gives a consistent estimator of the treatment effect under the sequential randomization assumption. Often, g-estimation is used along with a parameterized potential outcome model and a logistic model for the propensity score. However, the mle may not exist for some extreme treatment effect we plug in the model and this leads to some unidentifiability issues. We investigate and compare several methods for fixing this problem. Then we extend the discussion to the case of multiple treatments and relaxed assumptions (future ignorability assumption instead of sequential randomization assumption).

The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2012 program

2012 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.