The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
178
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2012 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section
|
Abstract - #305807 |
Title:
|
Unweighted General Linear Mixed Model
|
Author(s):
|
Byung Park*+ and Motomi Mori
|
Companies:
|
Oregon Health and Science University and Oregon Health and Science University
|
Address:
|
3181 SW Sam Jackson Park Road CR145, Portland, OR, 97239, United States
|
Keywords:
|
Unweighted means, General Linear Mixed Model, Hypothesis testing, Unweighted ANOVA
|
Abstract:
|
The unweighted means ANOVA (UANOVA) procedure was first described by Yates (1934) for an unbalanced two-way model with interaction as an approximate, but computationally simple method of analysis. Today the computationally simple part of a UANOVA justification is not particularly relevant because of high speed computers. Even so, the method is still sometimes recommended in today's literature. It is most popular in mixed linear models as a starting point for computing variance component estimators and confidence intervals. It has also been recommended as an alternative way for testing certain linear hypotheses in fixed linear models. It is the purpose of this study to investigate the UANOVA method for this less well developed area of testing linear hypotheses in fixed/Mixed effects models. Some properties of the unweighted means procedure are consistency of an unweighted means sum of squares and the distribution of that sum of squares. Our simulation study on fixed effect models showed that the power of the testing fixed effects in UANOVA appeared not to be sensitive to unbalancedness, while the power of testing using Type III sum of squares appeared to be more sensitive.
|
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 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.