|
Activity Number:
|
20
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #304884 |
|
Title:
|
Collaborative Targeted Maximum Likelihood Estimation
|
|
Author(s):
|
Susan Gruber*+ and Mark J. van der Laan
|
|
Companies:
|
University of California, Berkeley and University of California, Berkeley
|
|
Address:
|
c/o Biostatistics Dept., Berkeley, CA, 94720-7360,
|
|
Keywords:
|
causal inference ; TMLE ; machine learning ; epidemiology
|
|
Abstract:
|
We present a novel machine learning algorithm for estimation of a causal parameter that provides inference. Our two-stage approach is an extension of targeted maximum likelihood estimation methodology (van der Laan and Rubin, 2006) that incorporates super learning (Polley, Hubbard, van der Laan, 2007). Our method is data-adaptive, double-robust, and likelihood-based, employing cross validation to select among candidate estimators. It is readily implemented in standard software. Performance is illustrated using both simulated and real data.
|
- 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 |
