|
Activity Number:
|
364
- Contributed Poster Presentations: International Chinese Statistical Association
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
International Chinese Statistical Association
|
|
Abstract #314017
|
|
|
Title:
|
Local Signal Detection on Irregular Domain with Spatially Varying Coefficient Model
|
|
Author(s):
|
Chengzhu Zhang* and Lan Xue
|
|
Companies:
|
Oregon State University and Oregon State University
|
|
Keywords:
|
Irregular domain;
Oracle estimator;
Penalized least squares;
Bivariate spline;
SCAD;
Selection consistency
|
|
Abstract:
|
In some areas like spatial-temporal analysis and disaster analysis, a variety of questions require the understanding of spatial heterogeneity. They're required to accommodate coefficients dynamically change over special regions. Moreover, previous research work lacks consideration about detecting local signals. We study a model selection method for varying coefficient models by using penalized bivariate splines. It uses bivariate splines defined on triangulation to approximate nonparametric varying coefficient functions and minimizes the sum of squared errors with a local penalty on L2 norms of spline coefficients for each triangle. Our method partitions interested region into triangulation because using triangles to approximate curved boundaries is intuitive and efficient. For SCAD penalty, we propose an algorithm to obtain the estimator efficiently. We establish the consistency of the estimator and the estimated null region. The numerical performance of the proposed method is evaluated in both simulation case and real data analysis.
|
Authors who are presenting talks have a * after their name.
