|
Activity Number:
|
156
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #307434 |
|
Title:
|
Nonparametric Inference for High-Dimensional Longitudinal Data
|
|
Author(s):
|
Ke Zhang*+ and Haiyan Wang
|
|
Companies:
|
Kansas State University and Kansas State University
|
|
Address:
|
2212 Prairie Glen, Manhattan, KS, 66502,
|
|
Keywords:
|
high dimension ; longitudinal ; rank statistics ; lipidomics
|
|
Abstract:
|
Analysis of high dimensional data has received increasing attention in recent years as large amount of data become available due to fast development of modern technologies. In this paper, we consider testing the main effect of factors and the temporal change of the factor effects when there are a large number of levels for some factors and each subject is measured repeatedly over time. Test statistics are proposed based on both the original observations and their (mid-) ranks. The asymptotic distributions of the test statistics are obtained under corresponding null hypotheses. Simulation results will demonstrate the property of the tests. A lipidomics dataset are analyzed to illustrate the application of the proposed tests.
|
