在只有部份疾病訊息狀況下,有關二元診斷篩檢法之二個概似比特徵之概似推論方法;Likelihood methods for inference about the likelihood ratios of a binary diagnostic test in the presence of partial disease verification

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/86730

Title:	在只有部份疾病訊息狀況下,有關二元診斷篩檢法之二個概似比特徵之概似推論方法;Likelihood methods for inference about the likelihood ratios of a binary diagnostic test in the presence of partial disease verification
Authors:	劉亞幸;Liu, Ya-Hsing
Contributors:	數學系
Keywords:	概似函數;二元資料;驗證偏差;疾病訊息缺失;Likelihood ratio test;Binary data;Verification bias;Missing data
Date:	2021-07-28
Issue Date:	2021-12-07 13:09:52 (UTC+8)
Publisher:	國立中央大學
Abstract:	在醫學研究中，常使用二元診斷法來估計正概似比(positive likelihood ratio)與負概似比(negative likelihood ratio)，當遇到疾病訊息缺失時，要估計正概似比與負概似比往往相對困難些，本文針對疾病訊息是否有缺失時提出了概似函數方法。在獨立的伯努利概似函數推論下，可得到分數檢定、華德檢定與概似檢定。本文利用模擬與實例分析分別比較概似函數方法、Montero-Alonso與Roldan-Nofuentes (2018) 提出的華德檢定、Simel et al. (1991) 納皮爾概似比檢定與Walter (1975) 概似比檢定的型一誤差的機率、信賴區間上下界與覆蓋率，可發現本文的模型在解決疾病訊息缺失問題中更具優勢。;In medical research, the classic parameters use to assess the accuracy of a binary diagnostic test are the positive likelihood ratio and the negative likelihood ratio.The presence of missing data, When the disease information is missing, the positive likelihood ratio and the negative likelihood ratio are difficult to estimation.The article considers the disease information is missing or not, proposd the likelihood function method. The article use independent binary likelihood function, inferenced score test, wald test and likelihood ratios test statistics.We use simulation and real data analysis to demonstrate the merit of our new parametric robust technique. We also make comparison with to the wald test proposed by Montero-Alonso and Roldan-Nofuentes (2018), the Napierian likelihood test statistics proposed by Simel et al. (1991) and the likelihood test statistics proposed by Walter (1975).
Appears in Collections:	[Graduate Institute of Mathematics] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	81	View/Open

社群 sharing

Loading...