階層隨機效應模型的強韌性質之初探;Hierarchical random effects modeling - a robust perspective

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/74518

Title:	階層隨機效應模型的強韌性質之初探;Hierarchical random effects modeling - a robust perspective
Authors:	蘇建彰;Su, Chien-Chang
Contributors:	統計研究所
Keywords:	強韌概似函數;階層隨機效應模型;相關性群集資料;Robust likelihood;Hierarchical random effects model;clustered data
Date:	2017-06-20
Issue Date:	2017-10-27 14:00:22 (UTC+8)
Publisher:	國立中央大學
Abstract:	階層隨機效應模型常用來分析具有相關性的群集資料，而階層隨機效應模型組成分配的選擇常常是為了計算的方便。然而，本文的重點是在階層隨機效應模型的組合分配是否影響其最終的邊際分配具有可強韌化性質。我們將更換組成模型的分配，並探討邊際分配的迴歸參數之最大概似估計量和母體平均數之最大概似估計量是否具有一致性，推斷邊際分配是否具有可強韌化性質。 ;Hierarchical random effects models are convenient for modeling clustered data. More generally, model pecifications involved in the hierarchy are based on the ease of computation. We, on the other hand, focus on the issue of robustness. More specifically, we explore how the intermediate distributions in the hierarchy affect the robustness of the ultimate marginal distribution in providing consistent estimates for the population means/regression coefficients.
Appears in Collections:	[Graduate Institute of Statistics] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	297	View/Open

社群 sharing