Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/60177
Type: Artigo de periódico
Title: Influence diagnostics for the structural errors-in-variables model under the Student-t distribution
Author: Galea, M
Bolfarine, H
Vilca-Labra, F
Abstract: The influence of observations on the parameter estimates for the simple structural errors-in-variables model with no equation error, under the Student-t distribution, is investigated using the local influence approach. The main conclusion is that the Student-t model with small degrees of freedom is able to incorporate possible outliers and influential observations in the data. The likelihood displacement approach is useful for outlier detection, especially when a masking phenomenon is present and the degrees of freedom parameter is large. The diagnostics are illustrated with two examples.
Country: Inglaterra
Editor: Routledge Journals, Taylor & Francis Ltd
Rights: fechado
Identifier DOI: 10.1080/0266476022000011265
Date Issue: 2002
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
WOS000178356200008.pdf247.14 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.