|
Activity Number:
|
130
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
General Methodology
|
| Abstract - #305041 |
|
Title:
|
Weighted Likelihood for Semiparametric Models and Two-Phase Stratified Samples with Application to Cox Regression
|
|
Author(s):
|
Norman E. Breslow*+ and Jon A. Wellner
|
|
Companies:
|
University of Washington and University of Washington
|
|
Address:
|
Department of Biostatistics, Seattle, WA, 98195-7232,
|
|
Keywords:
|
case-cohort ; estimated weights ; failure time ; inverse probability weights ; missing data
|
|
Abstract:
|
Inverse probability weighted (IPW) likelihood equations lead to simple and robust estimators for two-phase stratified samples when used with semiparametric models having regular root N consistent estimators of Euclidean and nonparametric parameters. Under Bernoulli sampling, the influence function for the weighted likelihood estimator of the Euclidean parameter is the IPW version of the ordinary influence function. By proving weak convergence of the IPW empirical process and borrowing results on weighted bootstrap empirical processes, we show a parallel asymptotic expansion holds for finite population stratified sampling. Our key results were derived earlier for the special case of Cox regression with stratified case-cohort studies; other complex survey designs and missing data problems are discussed more generally. The paper interprets this prior work and paves the way for other applications.
|
