|
Activity Number:
|
565
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #303395 |
|
Title:
|
Numerical Algorithm-Based Estimation Methods for ODE Models
|
|
Author(s):
|
Arun Kumar*+ and Hulin Wu and Hongqi Xue
|
|
Companies:
|
University of Rochester and University of Rochester and Rochester University
|
|
Address:
|
6-1 Lilac Dr., Rochester, NY, 14620,
|
|
Keywords:
|
ordinary differential equation model ; HIV viral dynamics ; AIDS study
|
|
Abstract:
|
Differential equations are extensively used for modeling dynamics of physical processes in many scientific fields such as engineering, physics, and biomedical engineering. Estimating parameters of these models is a challenging problem because of higher computational cost for obtaining reasonably accurate estimate. In this talk, I will discuss a new method for estimating parameters of an ordinary differential equation (ODE) model. This new method exploits form of a numerical method for integrating an ODE to relate smoothed data with the model structure avoiding calculation of state variable's derivative through smoothing. The method improves upon performance of available quicker method in terms of accuracy of the estimate at a moderate increase in computational cost. Simulation results will be presented to illustrate method's effectiveness.
|
