|
Activity Number:
|
224
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304199 |
|
Title:
|
Weighted Fourier Analysis of Longitudinal Data
|
|
Author(s):
|
Shubing Wang*+ and Christopher Tong
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc.
|
|
Address:
|
RY 33-300, 126 East Lincoln Ave., Rahway, NJ, 07065,
|
|
Keywords:
|
Weighted Fourier Analysis ; Longitudinal Data ; Functional
|
|
Abstract:
|
When exploring and testing hypotheses of the treatment effect in a preclinical study or experiment, classic longitudinal data analysis methods, such as repeated measures ANOVA and ANCOVA, have many limitations when many time points or irregularly-spaced time points are observed, since it is difficult to describe the interactions of the time and treatment. Previous functional data analysis approaches have difficulty drawing correct comparative statistical inferences at local time points. We propose a method that estimates the unknown curves using a weighted Fourier series expansion. Weighted Fourier analysis characterizes the time and treatment interaction in a parsimonious form. Random field theory can be easily adapted for weighted Fourier analysis to detect local treatment differences properly for highly correlated longitudinal data.
|
- 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 2009 program |
