|
Activity Number:
|
291
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Council of Chapters
|
| Abstract - #303013 |
|
Title:
|
Inferring Causal Effects in High-Dimensional Studies
|
|
Author(s):
|
Mark J. van der Laan*+
|
|
Companies:
|
University of California, Berkeley
|
|
Address:
|
108 Haviland Hall, School of Public Health, Berkeley, CA, 94720-7360,
|
|
Keywords:
|
|
|
Abstract:
|
We present a general targeted maximum likelihood based approach targeting a user supplied variable importance or causal effect on an outcome of interest. This method provides an optimal way in dealing with the curse of dimensionality while preserving the likelihood as principle criterion. We illustrate this method for the purpose of assessing the effect of mutations in the HIV virus that cause resistance to a particular drug, and for detecting binding sites in the regulatory region of the yeast genome based on gene expression experiments. We also apply the targeted maximum likelihood methodology to case control data providing a new way of assessing causal effects of single nucleotide mutations on phenotypes of interest.
|
- 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 |
