• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.11.2024.tde-04042024-093933
Document
Author
Full name
Welinton Yoshio Hirai
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Piracicaba, 2023
Supervisor
Committee
Piedade, Sônia Maria De Stefano (President)
Biase, Adriele Giaretta
Dias, Carlos Tadeu dos Santos
Hongyu, Kuang
Title in Portuguese
Estudo do método de estimação do tipo razão multivariada para média populacional
Keywords in Portuguese
Amostragem
Correlação
Estimador pela razão
Imputação
Abstract in Portuguese
Nesse trabalho foram produzidos estudos para um método de estimação da média baseado em mais que uma variável auxiliar, o método da razão multivariada. A abordagem utiliza-se da razão entre a variável de interesse em conjunto com demais variáveis coletadas durante o processo de amostragem para aprimorar o valor preditivo da estimativa. No primeiro capítulo, foi realizada uma comparação entre o método usual, método da razão univariada, e o método da razão multivariada para cenários simulados de alta e baixa correlação entre grupos. No segundo capítulo, foi proposto um novo método de imputação utilização a razão multivariada, estudo das expressões matemáticas para boas propriedades e aplicados em dados simulados e reais. O método se apresenta com bons resultados para estruturas altamente correlacionadas. Os códigos foram implementados na linguagem R e anexadas no trabalho.
Title in English
Study of the multivariate ratio estimation method for population mean
Keywords in English
Correlation
Imputation
Ratio estimation
Sampling
Abstract in English
In this work, studies were conducted to develop a mean estimation method based on more than one auxiliary variable, known as the multivariate ratio method. The approach utilizes the ratio between the variable of interest and other variables collected during the sampling process to enhance the predictive value of the estimate. In the first chapter, a comparison was made between the usual method, the univariate ratio method, and the multivariate ratio method in simulated scenarios with high and low correlation between groups. In the second chapter, a new imputation method using the multivariate ratio was proposed, with a study of mathematical expressions to ensure good properties and application to simulated and real data. The method showed good results, especially in highly correlated structures. The codes were implemented in the R language and are attached to the work.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2024-04-04
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.