|
Activity Number:
|
343
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #306033 |
|
Title:
|
Measuring the Mean Squared Error of the EBLUP in Linear Mixed Models
|
|
Author(s):
|
Jamie McClave Baldwin*+ and Ramon Littell
|
|
Companies:
|
Info Tech, Inc. and University of Florida
|
|
Address:
|
5700 SW 34th Street, Suite 1235, Gainesville, FL, 32608,
|
|
Keywords:
|
linear mixed models ; EBLUP ; MSEP ; Kenward-Roger adjustment ; propagated variance
|
|
Abstract:
|
In 1984, Kackar and Harville (JASA, 79) introduced a new method to handle the propagation of variance stemming from unknown variance parameters in linear mixed models. In the 20+ years since this development there have been many adjustments, expansions and new techniques stemming from their idea. In this paper, we review the (frequentist) procedures now available for adjusting the Mean Squared Error of Prediction (MSEP) of the EBLUP in the setting of Linear Mixed Models to account for estimating unknown covariance parameters. While much has been done to improve the estimation of the MSEP, many analysts do not understand when or why it is important to make these corrections or even what the corrections are. This paper attempts to answer the questions: What is the precision of the EBLUP? How can we measure it? When and how should we adjust for the extra variation?
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
