|
Activity Number:
|
523
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #305655 |
|
Title:
|
Functional Latent Feature Models for Data with Longitudinal Covariate Processes
|
|
Author(s):
|
Erning Li*+ and Yehua Li and Nae-Yuh Wang and Naisyin Wang
|
|
Companies:
|
Texas A&M University and The University of Georgia and Johns Hopkins University School of Medicine and Texas A&M University
|
|
Address:
|
Department of Statistics, College Station, TX, 77843-3143,
|
|
Keywords:
|
Asymptotics ; Kernel method ; Longitudinal data ; Measurement error ; Functional principal component analysis ; Sufficiency score
|
|
Abstract:
|
We consider a joint model approach to study the association between nonparametric latent features of longitudinal processes and a primary endpoint. We argue that the key assumption in the usual generalized functional linear models that the estimation variation in eigenfunctions is negligible for all eigen-components is not necessarily true and is purely determined by the nature of data. We propose estimation procedures and supporting theory that allow investigations regardless the validity of this key assumption. Our approach takes into account the estimation uncertainty embedded in the estimated eigen-system and allows users to have a thorough understanding of where the estimation uncertainty/variation lies so that the choice of a final model and the future research plan can be made accordingly. The proposed method is evaluated through simulations and via a hypertension data analysis.
|
