|
Activity Number:
|
582
- Statistical Methods for Functional Data
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 2, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
|
Abstract #323875
|
View Presentation
|
|
Title:
|
Optimal Design for Classification of Functional Data
|
|
Author(s):
|
Cai Li* and Luo Xiao
|
|
Companies:
|
North Carolina State University and North Carolina State University
|
|
Keywords:
|
Classification ;
Covariance function ;
Design point ;
Linear discriminant analysis ;
Longitudinal Data ;
Functional data
|
|
Abstract:
|
We consider the problem of selecting sampling points for optimal classification of functional data. We propose optimal designs that are appropriate to both dense functional data and sparse functional data. Our method is based on linear discriminant analysis. We formulate explicit design objectives as functions of sampling points. Moreover, we study the theoretical properties of the proposed design objectives and provide a computationally simple implementation. The performance of our method is assessed through extensive numerical studies, including simulations and various real data applications.
|
Authors who are presenting talks have a * after their name.
