National Changhua University of Education Institutional Repository : Item 987654321/14055

English | 正體中文 | 简体中文 | Items with full text/Total items : 6507/11669
Visitors : 30392890 Online Users : 470

RC Version 3.2 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.

Scope Adv. Search

Login ‧ Upload ‧ Help ‧ About ‧ Administer

NCUEIR > College of Science > ma > NSC Projects > Item 987654321/14055

Please use this identifier to cite or link to this item: http://ir.ncue.edu.tw/ir/handle/987654321/14055

Title:	空間迴歸模型變數選取方法(1/2) Variable Selection for Spatial Regression Models
Authors:	陳春樹
Contributors:	統計資訊研究所
Keywords:	Conditional information criterion;Data perturbation;Spatial prediction;Stabilization;Stein’s unbiased risk estimate;Variable selection.
Date:	2009
Issue Date:	2012-09-10T04:40:17Z
Publisher:	行政院國家科學委員會
Abstract:	Variable selection in geostatistical regression is an important problem, but has not been well studied in the literature. In the first year of this project, we focus on spatial prediction and consider a class of conditional information criteria indexed by a penalty parameter. Instead of directly applying a criterion, which leads to a discontinuous spatial predictor with respect to the response variables, we further stabilize the predictor by local model averaging, resulting in a predictor that is not only continuous but also differentiable. Then Stein’s unbiased risk estimate is applied to select the penalty parameter, which takes estimation of model parameters into account and can be computed using a simple Monte Carlo method. In the second year of this project, we will continue to complete the expected goals.
Relation:	國科會計畫, 計畫編號: NSC98-2118-M018-003-MY2; 研究期間: 9808-9907
Appears in Collections:	[ma] NSC Projects

Files in This Item:

File	Size	Format
2020800312001.pdf	93Kb	Adobe PDF	442	View/Open