• 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.11.1999.tde-20210104-192558
Document
Author
Full name
Sílvia Helena Venturolli Perri
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Piracicaba, 1998
Supervisor
Title in Portuguese
Ajuste de modelos mistos de classificação dupla: procedimentos do sistema estatístico "SAS"
Keywords in Portuguese
COMPONENTES DE VARIÂNCIA
MODELOS MISTOS
SAS (SOFTWARE ESTATÍSTICO)
TESTES DE HIPÓTESES
Abstract in Portuguese
O Sistema Estatístico SAS apresenta grande flexibilidade para o ajuste de modelos mistos através dos procedimentos abordados GLM, V ARCOMP e MIXED, destacando-se a excelente performance do procedimento MIXED. Embora o PROC MIXED seja o procedimento específico para a análise de modelos mistos, o PROC GLM com a declaração RANDOM e a opção TEST e o PROC VARCOMP com a opção FIXED, também podem ser utilizados. Uma das vantagens do PROC MIXED consiste em fornecer através da declaração RANDOM e da opção TYPE, várias estruturas para a matriz de variâncias e covariâncias, enquanto que nos procedimentos GLM e V ARCOMP, está disponível apenas a estrutura de componentes de variância (VC). Neste estudo foram adotadas as estruturas VC, CS, TOEP(l) e HF, dentre as quais, segundo os critérios AIC e BIC, apenas a estrutura HF não se mostrou adequada. Este fato foi também observado na estimação dos componentes de variância e nos testes de hipóteses sobre o fator de efeitos fixos.
Title in English
Fitting mixed models for the two-way layout: procedures of the statistical software "SAS"
Keywords in English

Abstract in English
The Statistical Software SAS shows large flexibility of fitting rnixed models from the GLM, VARCOMP and MIXED procedures, distinguishing the excellent performance of the MIXED procedure. Although the PROC MIXED is the specific procedure for the analysis of mixed models, the PROC GLM with the RANDOM statement and the TEST option and the PROC VARCOMP with the FIXED option can also be used. One of the advantage of the PROC MIXED is to provide with the RANDOM statement and the TYPE option, many structures for the variance and covariance matrix, while in the GLM and VARCOMP procedures, only one structure of variance components (VC) is available. ln this work, the VC, CS, TOEP(l) and HF structures were used, among them, according to AIC and BIC criteria, only the HF structure was not appropriate. This fact was also observed in the variance components estimators and in the hypothesis tests on the fixed effects factor.
 
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
2021-01-07
 
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.