淡江大學機構典藏:Item 987654321/87474
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 64198/96992 (66%)
造訪人次 : 7967649      線上人數 : 9381
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/87474


    題名: 具共變量測量誤差之現狀數據分析
    其他題名: Analysis of current status data with mismeasured covariate
    作者: 傅婕寧;Fu, Chieh-Ning
    貢獻者: 淡江大學數學學系碩士班
    温啓仲;Wen, Chi-Chung
    關鍵詞: 條件分數;現狀數據;測量誤差;比例勝算比模型;右設限;篩選法;Conditional score;Current Status Data;measurement error;Proportional odds model;Right-censored;Sieve
    日期: 2012
    上傳時間: 2013-04-13 11:11:48 (UTC+8)
    摘要: 雖然已有相當多的研究工作在右設限存活數據具共變量測量誤差的問題,但具共變量測量誤差現狀數據的迴歸分析近年來才開始。這篇論文中,我們考慮Wen and Chen (2012) 分析具共變量誤差現狀數據的條件分數法。不同於Wen and Chen (2012),我們使用篩選法去近似比例勝算比模型的基線勝算比函數。我們所提方法的優點為條件分數估計的計算更有效率且迴歸係數估計之變異數估計無需估計設限時間分布。所提的方法藉由模擬試驗評量且由一實際數據說明其應用。
    While there is considerable work on covariate measurement error problem for right-censored survival data, regression analysis for current status data with mismeasured covariate has just started recently. In this thesis, we consider the conditional score approach of Wen and Chen (2012) for the analysis of errors-in-variable current status data. Distinct from Wen and Chen (2012), we employ a sieve method to approximate the baseline odds ratio function in the proportional odds model. Our proposal has advantages that the computation of conditional score estimates is more efficient and the variance estimation of regression parameter estimate is exempted from estimating the distribution of censoring time. The proposed method is evaluated through simulation studies and illustrated with a real data.
    顯示於類別:[應用數學與數據科學學系] 學位論文

    文件中的檔案:

    檔案 大小格式瀏覽次數
    index.html0KbHTML271檢視/開啟

    在機構典藏中所有的資料項目都受到原著作權保護.

    TAIR相關文章

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