|
Activity Number:
|
124
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 1, 2016 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
|
Abstract #320242
|
View Presentation
|
|
Title:
|
Significance Tests for Time-Varying Covariate Effect in Longitudinal Functional Data
|
|
Author(s):
|
Saebitna Oh* and Ana-Maria Staicu
|
|
Companies:
|
North Carolina State University and North Carolina State University
|
|
Keywords:
|
Longitudinal data ;
Functional data ;
Pseudo F test
|
|
Abstract:
|
We consider time-varying functional regression models to describe associations between longitudinal functional responses, where functions are observed at multiple instances (often visit times) per subject for many subjects, and subject specific covariates. We develop inferential methods to assess the significance of the time-varying covariate effect. We propose a pseudo F- testing procedure that accounts for the complex error structure and is computationally efficient. Numerical studies confirm that the testing approach has the correct size and compares favorably with available competitors in terms of power. The methods are illustrated on a data application.
|
Authors who are presenting talks have a * after their name.
