|
Activity Number:
|
275
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #302392 |
|
Title:
|
Fitting Stratified Proportional Odds Models by Amalgamating Conditional Likelihoods
|
|
Author(s):
|
Bhramar Mukherjee and Jaeil Ahn*+
|
|
Companies:
|
The University of Michigan and The University of Michigan
|
|
Address:
|
M4153, Ann Arbor, 48109-2029,
|
|
Keywords:
|
conditional likelihood ; nuisance ; proportional odds ; sandwich estimate
|
|
Abstract:
|
Classical methods for fitting a varying intercept logistic regression model to stratified data are based on the conditional likelihood principle to eliminate the stratum-specific nuisance parameters. However, classical conditioning techniques do not apply to the general K-category cumulative logit model (K>2) with varying stratum-specific intercepts and there is no reduction due to sufficiency; the nuisance parameters remain even in the conditional likelihood. We propose a methodology to fit stratified proportional odds model in a general regression set-up, by amalgamating conditional likelihoods. We provide a robust sandwich estimate of the variance of the proposed estimator. Simulation studies comparing the proposed method with a random effects model on the stratification parameter are also furnished.
|
