|
Activity Number:
|
284
- GMM, Triple Joint Modeling, Bootstrapping, and Multiple Membership of Correlated Data
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 1, 2017 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
WNAR
|
|
Abstract #323644
|
|
|
Title:
|
GMM Logistic Regression with Time-Dependent Covariates and Feedback Processes
|
|
Author(s):
|
Kyle M Irimata*
|
|
Companies:
|
Arizona State University
|
|
Keywords:
|
generalized linear models ;
correlated data ;
hierarchical data ;
moment conditions ;
longitudinal data
|
|
Abstract:
|
This is the first of four related papers addressing the intraclass correlation due to the hierarchical structure of the data. The analysis of longitudinal data requires a model which accounts for both the inherent correlation amongst the responses as a result of the repeated measurements, as well as the feedback between the responses and any time-dependent covariates. Lalonde, Wilson and Yin (2013) developed an approach based on generalized methods of moments (GMM) for identifying and using valid moment conditions to account for time dependent covariates in longitudinal data. However, this model does not provide information regarding the specific relationship that exists across time-points between the response and covariates. In particular, the effect of a given predictor on the response may vary across time through either fall off or lagged effects. We extend the GMM approach discussed by Lalonde, Wilson and Yin to evaluate the relationship and potential feedback between the response and the covariates at different periods of time using valid moment conditions.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2017 program
|
