English  |  正體中文  |  简体中文  |  Items with full text/Total items : 58335/91896 (63%)
Visitors : 29732      Online Users : 85
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/103000

    Title: 遺失資料結構對插補策略應用在不完全長期追蹤資料之影響
    Other Titles: The Impact of Missing-Data Mechanisms on Imputation Strategies for Incomplete Longitudinal Data
    Authors: 陳怡如
    Contributors: 淡江大學統計學系
    Keywords: 不完全長期追蹤資料;非隨機遺失;多重插補法incomplete longitudinal data;not missing at random;multiple imputation
    Date: 2012-08
    Issue Date: 2015-05-13 11:19:28 (UTC+8)
    Abstract: 遺失資料常發生於長期追蹤研究中,多重插補即為其中一種解決遺失值問題的有效方法。在初期時,多重插補方法主要應用於抽樣調查與普查方面,然而近年來持續擴展應用於生物醫學、行為與社會科學等領域。雖然已有些學者提出處理有關不完全長期追蹤二元資料與不完全長期追蹤順序資料的多重插補法,但是鮮少有討論在不同遺失資料結構下,這些插補策略之表現。遺失資料結構之重要性,在於處理遺失資料方法之特性極度仰賴遺失資料結構。遺失資料結構包含完全隨機遺失、隨機遺失與非隨機遺失。一般而言,實務上較會發生遺失值的情況屬於非隨機遺失。在之前的研究計畫中,已應用模擬研究來探討一些插補策略,就標準化偏誤、均方根誤差與涵蓋率等衡量指標,以瞭解在完全隨機遺失和隨機遺失下之表現。此申請研究計畫之主要研究目標,則是更進一步討論現有的插補策略與研究中的新插補策略,在非隨機遺失情況下之表現。
    Multiple imputation method is one of the effective approaches to solve the problem of missing data which are commonly occurred in longitudinal studies. The initial use of multiple imputation is primarily for sample surveys and censuses. Nowadays, this attractive approach has been increasingly utilized in the biomedical, behavioral, and social sciences. Several imputation strategies have been proposed for dealing with incomplete longitudinal binary or ordinal data. However, few of them discuss the influence of missing-data mechanisms on those imputation strategies. The main reason that missing-data mechanisms are crucial is the properties of missing-data approaches depending very strongly on the mechanisms. Missing-data mechanisms include missing completely at random (MCAR), missing at random (MAR) and not missing at random (NMAR). In the previous project, the performances of some imputation strategies have been evaluated in terms of standardized bias, root-mean-squared error and coverage percentage under MCAR and MAR. The aim of this project is to further investigate the impact of NMAR on the current imputation strategies and the ongoing approach for incomplete longitudinal data, which is more practical in real-life applications.
    Appears in Collections:[Graduate Institute & Department of Statistics] Research Paper

    Files in This Item:

    There are no files associated with this item.

    All items in 機構典藏 are protected by copyright, with all rights reserved.

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