|
Activity Number:
|
106
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
ENAR
|
| Abstract - #309192 |
|
Title:
|
Overdispersion in Multilevel Correlated Data
|
|
Author(s):
|
Trent Lalonde*+ and Jeffrey R. Wilson
|
|
Companies:
|
Arizona State University and Arizona State University
|
|
Address:
|
Department of Mathematics and Statistics, Tempe, AZ, 85287,
|
|
Keywords:
|
quasilikelihood ; censoring ; heterogeneity ; extravariation ; random effects
|
|
Abstract:
|
A common difficulty in regression problems with the exponential family response variables is overdispersion. While complex designs such as a multistage stratified cluster design greatly enhances the feasibility of data collection, it results in differential probabilities of selection as would not be the case if simple random sampling approach was employed. This phenomenon known as overdispersion has been examined through different approaches by several researchers. However, most articles either account for overdispersion caused by lack of independence among units or variation among responses within groups. This paper examines random effect mixed models that accounts simultaneously as well as at multilevels data.
|
