|
Activity Number:
|
360
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #301972 |
|
Title:
|
ANOVA for Semiparametric Models with Missing and Unbalanced Longitudinal Data
|
|
Author(s):
|
Pingshou Zhong*+ and Song Xi Chen
|
|
Companies:
|
Iowa State University and Iowa State University
|
|
Address:
|
210 Snedecor Hall, Ames, IA, 50010,
|
|
Keywords:
|
ANOVA ; Longitudinal data ; Empirical likelihood
|
|
Abstract:
|
We consider ANOVA analyses for a semi-parametric partially linear regression model with missing and unbalanced longitudinal data from different treatment groups. We assume the missing is monotone and the propensity function is known up to a parameter. Empirical likelihood ratio statistics are formulated for both the equality of the parametric regression parameters and the nonparametric regression functions. It is found that the asymptotic distributions of these statistics, respectively, are chi-square and normal distribution, which lead to readily implementable nonparametric likelihood ratio ANOVA tests for both the parametric and nonparametric parts of the regression models. The performance of the ANOVA tests are evaluated empirically by simulation and a case study.
|
