|
Activity Number:
|
152
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #306071 |
|
Title:
|
Instrumental Variable Estimation in Logistic Regression Models with Measurement Error
|
|
Author(s):
|
Kimberly Weems*+ and Leonard A. Stefanski
|
|
Companies:
|
North Carolina State University and North Carolina State University
|
|
Address:
|
Department of Statistics, 220 D Patterson Hall, Raleigh, NC, 27695,
|
|
Keywords:
|
logistic regression ; measurement error ; instrumental variables
|
|
Abstract:
|
We consider parameter estimation in logistic regression models with measurement error using a binary instrumental variable. The conditional-score method of Stefanski and Carroll (1987) is used to obtain unbiased estimating equations. We obtain sufficient statistics for the unobserved predictors and the conditional distribution of the observed data given these sufficient statistics. Unbiased score functions that are free of the unknown predictors are then used to derive unbiased estimating equations for the model parameters. Our work generalizes that of Buzas and Stefanski (1996) to non-normal instrumental variables. Simulation results and an application to real data are presented to illustrate this method.
|
