2026-03-15T14:09:35Z https://doshisha.repo.nii.ac.jp/oai

oai:doshisha.repo.nii.ac.jp:00029233 2025-07-11T06:38:38Z 4151:8699:4185:4186 4251:7158:7202:8771:7203

構造的潜在変数モデリングに基づく複雑な線形回帰 コウゾウテキ センザイ ヘンスウ モデリング ニモトズク フクザツナ センケイ カイキ Complex linear regression based on structural latent variable modeling Binh An, Duong Thi 30775 9000399220032 構造潜在変数 線形回帰 潜在変数 structural latent variable linear regression latent variable The content discusses the importance of models in problem-solving, focusing on linear models. Linear models' strengths lie in their simplicity and applicability, yet they require modifications for complex real-world issues. The dissertation aims to contribute to multivariate linear model literature with two key proposals. Chapter 2 introduces a modified Deming Regression, dealing with multiple variables and errors. Chapter 3 delves into Structural Equation Modeling (SEM), addressing its challenges and proposing a K-means Generalized Maximum Entropy method. This enhances SEM's efficiency by using k-means centroids. Each chapter is outlined, detailing their significance and methodologies, and simulation results are presented. The research strives for improved modeling approaches in practical applications. doctoral thesis 2022-09-17 application/pdf application/pdf 甲第1247号 https://doshisha.repo.nii.ac.jp/record/29233/files/k1247.pdf https://doshisha.repo.nii.ac.jp/record/29233/files/zk1247.pdf https://doi.org/10.14988/0000029233 https://doshisha.repo.nii.ac.jp/records/29233 eng BD04086069 掲載刊行物所蔵情報へのリンク ／ Link to Contents https://doors.doshisha.ac.jp/opac/opac_link/bibid/BB13219763/?lang=0 open access