|
Activity Number:
|
274
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #302567 |
|
Title:
|
Self-Modeling Regression with Regression Splines and Random Curve-Specific Parameters: Application to Arterial Pulse Pressure Waveforms
|
|
Author(s):
|
Lyndia C. Brumback*+ and Douglas Tommet and Richard Kronmal
|
|
Companies:
|
University of Washington and Institute for Aging Research and University of Washington
|
|
Address:
|
Box 357232, Seattle, WA, 98115,
|
|
Keywords:
|
functional data ; nonlinear mixed effects models ; pulse waveforms ; self-modeling
|
|
Abstract:
|
Self modeling regression is a method for analyzing sets of observed curves. It is based on the relatively simple assumption that the x and y axes can be separately transformed in a parametric manner for each curve so that the data from all curves lie approximately on one typical curve. When the typical curve is modeled with a regression spline and the curve-specific transformational parameters are modeled as random (Normal with mean zero), the model may be under-parameterized and the variance components may be estimated poorly. Simulations show that a random effects distribution that forces the realized curve-specific transformational parameters to have mean zero or the inclusion of a fixed transformational parameter improves estimation. The methods are applied to arterial pulse pressure waveform data from the Multi-Ethnic Study of Atherosclerosis.
|
