|
Activity Number:
|
45
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #303128 |
|
Title:
|
The Analysis of Retrospective Family Studies
|
|
Author(s):
|
John Neuhaus*+ and Alastair Scott and Chris Wild
|
|
Companies:
|
University of California, San Francisco and University of Auckland and University of Auckland
|
|
Address:
|
Division of Biostatistics, San Francisco, CA, 94107-1762,
|
|
Keywords:
|
Case-control family study ; Responsive-selective sampling ; Retrospective sampling
|
|
Abstract:
|
Genetic epidemiologists often augment an initial case-control sample with responses and covariates gathered from the family members of the initial sample persons (probands) to assess within-family covariate effects and measure familial aggregation (within-family dependence) of the response. Standard likelihoods for such response selective samples typically involve a potentially infinite nuisance dimensional parameter, the marginal distribution of the covariates. Since we would like to avoid modeling this distribution, we develop semiparametric methods which treat the distribution of the covariates nonparametrically. This talk describes our approach and illustrates it using data from several case-control family studies of cancer.
|
