Activity Number:
|
131
|
Type:
|
Invited
|
Date/Time:
|
Monday, August 12, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
ENAR
|
Abstract - #300070 |
Title:
|
The Use of Polynomial Splines in Longitudinal Data Analysis
|
Author(s):
|
Jianhua Huang*+
|
Affiliation(s):
|
University of Pennsylvania
|
Address:
|
The Wharton School, Philadelphia, Pennsylvania, 19104, USA
|
Keywords:
|
polynomial spline ; nonparametric estimation ; repeated measurements
|
Abstract:
|
Nonparametric regression models are flexible and impose less restrictive assumptions than parametric models. However, fully nonparametric models can not be adequately estimated under most realistic sample sizes when there are many covariates due to the well-known "curse of dimensionality." It is necessary to introduce some structures to the regression function to make the estimation problem practically feasible. In this talk, we will show that polynomial splines provide a convenient tool for estimation for structured nonparametric models. In particular, we will discuss in some detail a time-varying coefficient model, a natural structured nonparametric model for longitudinal data analysis.
|
- 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 2002 program |