|
Activity Number:
|
91
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #303197 |
|
Title:
|
The Impact of the Winner's Curse on Parameter Estimation and False Positive Rates in Studies Involving Environmental Exposure Data
|
|
Author(s):
|
Sebastian Zöllner*+
|
|
Companies:
|
University of Michigan
|
|
Address:
|
, , ,
|
|
Keywords:
|
Winner's Curse ; GWAS
|
|
Abstract:
|
As researchers begin to incorporate environmental exposure data into association studies it is important to understand how the Winner's Curse will impact parameter estimation and false positive rates studies including gene-environment interaction. We explore this question by simulating case-control data sets under gene-environment interaction and test for genetic association. For markers significantly associated with the phenotype, we estimate its effect size and gene-environment interaction parameters. In the presence of main and interaction effects we observe overestimates of effect with bias increasing as power of the initial test decreases. More surprising is the potential for the Winner's Curse to induce false positives. We also present strategies for overcoming the Winner's curse to generate unbiased estimates of effect sizes and well calibrated tests of GxE interaction.
|
- 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 |
