English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 62568/95225 (66%)
造访人次 : 2510310      在线人数 : 213
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜寻范围 查询小技巧:
  • 您可在西文检索词汇前后加上"双引号",以获取较精准的检索结果
  • 若欲以作者姓名搜寻,建议至进阶搜寻限定作者字段,可获得较完整数据
  • 进阶搜寻

    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/91870

    题名: Functional Data Classification via Covariate Adjusted Subspace Projection
    作者: LI, PAI-LING
    贡献者: 淡江大學統計學系
    日期: 2013-07-24
    上传时间: 2013-07-30 19:48:02 (UTC+8)
    摘要: A covariate adjusted subspace projected functional data classification (SPFC) method is proposed
    for curves or functional data classification with accommodating additional covariate information.
    Based on the framework of subspace projected functional data clustering, curves of each cluster are
    embedded in the cluster subspace spanned by a mean function and eigenfunctions of the covariance
    kernel. We assume that the mean function may depend on covariates, and curves of each cluster
    are represented by the covariate adjusted functional principal components analysis (FPCA) model or
    covariate adjusted Karhunen-Loève expansion. Under the assumption that all the groups have different
    mean functions and eigenspaces, an observed curve is classified into the best predicted class by
    minimizing the distance between the observed curve and predicted functions via subspace projection
    among all clusters based on the covariate adjusted FPCA model. The proposed covariate adjusted
    SPFC method that accommodates additional information of other covariates is advantageous to improving
    the classification error rate. Numerical performance of the proposed method is examined by
    simulation studies, with an application to a data example.
    關聯: The 29th European Meeting of Statisticians
    显示于类别:[統計學系暨研究所] 會議論文





    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回馈