|
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 #323638
|
|
|
Title:
|
Joint Modeling of Mean, Dispersion, and Weights
|
|
Author(s):
|
Katherine Irimata*
|
|
Companies:
|
Arizona State University
|
|
Keywords:
|
joint modeling of the mean and variance ;
double generalized linear models ;
correlated data ;
overdispersion ;
generalized linear models
|
|
Abstract:
|
This is the second of four related papers addressing the intraclass correlation due to the hierarchical structure of the data. While generalized linear models have additional flexibility due to the link function, the variance is often constrained as a function of the mean. However, the joint modeling of the mean and the dispersion accounts for additional variation that is not explained by the mean submodel. This additional variation is modeled through covariates, Smyth (1989). This technique uses the deviance from the mean model as the dependent variable for the dispersion model. We expand on this approach to jointly model the weights. We discuss the benefits of incorporating the weights through a simulation study and an application to a numerical example.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2017 program
|
