Activity Number:
|
403
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 15, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Bayesian Stat. Sciences*
|
Abstract - #300759 |
Title:
|
Analysis of Growth Curves via Mixtures
|
Author(s):
|
Ori Rosen*+ and Ayala Cohen
|
Affiliation(s):
|
University of Pittsburgh and Technion-Israel Institute of Technology
|
Address:
|
2702 Cathedral of Learning, Pittsburgh, Pennsylvania, 15260, U.S.A.
|
Keywords:
|
Growth Curves ; Mixtures-of-Experts
|
Abstract:
|
In this paper we analyze a large collection of growth curves of male and female infants consisting of height measurements as a function of age. We propose a two-stage model consisting of first fitting the individual growth curves and then estimating the population parameters via a mixed effects model. In the first stage, we use mixtures-of-experts, originally introduced in the neural network literature. This model does not assume a specific shape for the growth curves and is thus more flexible than standard models.
|