|
Activity Number:
|
295
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #302752 |
|
Title:
|
Causal Inference for Continuous Time Longitudinal Data When Covariates Are Observed Only at Discrete Times
|
|
Author(s):
|
Dylan Small*+ and Mingyuan Zhang and Marshall Joffe
|
|
Companies:
|
University of Pennsylvania and University of Pennsylvania and University of Pennsylvania
|
|
Address:
|
400 Huntsman Hall, 3730 Walnut St., Philadelphia, PA, 19104,
|
|
Keywords:
|
causal inference ; continuous time process ; g-estimation ; longitudinal data
|
|
Abstract:
|
Most of the work on g-estimation for causal inference in longitudinal data assumes a discrete time underlying data generating process. However, in some studies, it is more reasonable to assume that the data are generated from a continuous time process, but the covariates are only observable at discrete times. For this setting, we study the assumptions needed for discrete time g-estimation to provide consistent estimates and present a new method that provides consistent estimates under weaker assumptions than usual discrete time g-estimation. We use our new method to study the effect of diarrhea on children's height, using a data set collected following a massive flood in Bangladesh.
|
- 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 2009 program |
