長期追蹤資料上的 Gamma-EM 分群;Gamma-EM clustering on longitudinal data

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/89629

Title:	長期追蹤資料上的 Gamma-EM 分群;Gamma-EM clustering on longitudinal data
Authors:	李宥序;Li, Yu-Hsu
Contributors:	統計研究所
Keywords:	EM 演算法;散度;長期追蹤資料;線性混合模型;分群;EM algorithm;divergence;longitudinal data;linear mixed effect model;clustering
Date:	2022-08-01
Issue Date:	2022-10-04 11:50:00 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨著現代科技和醫學的進步，已經有很多精密儀器可以準確檢測各種生物指標。在實踐中，研究一種藥物是否具有顯著效果是藥物研發中的一個重要問題。傳統上，我們會驗證實驗組和對照組是否有顯著差異，然後解釋藥物療效是否有效；然而，在一些臨床數據上，我們不知道數據背後的分群，進而判斷藥物有效性，本文將採用 PBC 資料作為例子。這裡我們使用Lin 和Wang（2021）提出的γ-EM 算法對未知群體的種群進行聚類分析。γ-EM 是通過γ-divergence 改進的EM 算法，可用於實現分類的魯棒性。在這種情況下，我們可以使用γ-EM 來初步了解種群是否具有不同群體的表現。;With the advancement of modern technology and medicine, there are already many sophisticated instruments that can accurately detect various biological indicators. In practice, it is an important issue in drug research and development to study whether a drug has a significant effect. Traditionally, we will verify whether there is a significant difference between the experimental group and the control group, and then explain whether the drug efficacy is effective; from another perspective, here we use the γ-EM algorithm proposed by Lin and Wang(2021) to perform cluster analysis on the population of unknown groups. γ-EM is an improved EM algorithm through γ-divergence, which can be used to achieve robustness in classification. In this case, we can use γ-EM to initially understand whether the population has the performance of different groups.
Appears in Collections:	[Graduate Institute of Statistics] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	124	View/Open

社群 sharing

Loading...