Generalized ridge estimators adapted in structural equation models

Detalhes bibliográficos
Autor(a) principal: Pereira, Gislene Araujo
Data de Publicação: 2020
Outros Autores: Resende, Mariana, Cirillo, Marcelo Ângelo
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Acta scientiarum. Technology (Online)
Texto Completo: http://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/49929
Resumo: Multicollinearity is detected via regression models, where independent variables are strongly correlated. Since they entail linear relations between observed or latent variables, the structural equation models (SEM) are subject to the multicollinearity effect, whose numerous consequences include the singularity between the inverse matrices used in estimation methods. Given to this behavior, it is natural to understand that the suitability of these estimators to structural equation models show the same features, either in the simulation results that validate the estimators in different multicollinearity degrees, or in their application to real data. Due to the multicollinearity overview arose from the fact that the matrices inversion is impracticable, the usage of numerical procedures demanded by the maximum likelihood methods leads to numerical singularity problems. An alternative could be the use of the Partial Least Squares (PLS) method, however, it is demanded that the observed variables are built by assuming a positive correlation with the latent variable. Thus, theoretically, it is expected that the load signals are positive, however, there are no restrictions to these signals in the algorithms used in the PLS method. This fact implies in corrective areas, such as the observed variables removal or new formulations of the theoretical model. In view of this problem, this paper aimed to propose adaptations of six generalized ridge estimators as alternative methods to estimate SEM parameters. The conclusion is that the evaluated estimators presented the same performance in terms of accuracy, precision while considering the scenarios represented by model without specification error and model with specification error, different levels of multicollinearity and sample sizes.
id UEM-6_56c06e036201a07ecd61db73ffa37962
oai_identifier_str oai:periodicos.uem.br/ojs:article/49929
network_acronym_str UEM-6
network_name_str Acta scientiarum. Technology (Online)
repository_id_str
spelling Generalized ridge estimators adapted in structural equation modelsGeneralized ridge estimators adapted in structural equation modelsstructural model; generalized ridge regression; multicollinearity.structural model; generalized ridge regression; multicollinearity.Multicollinearity is detected via regression models, where independent variables are strongly correlated. Since they entail linear relations between observed or latent variables, the structural equation models (SEM) are subject to the multicollinearity effect, whose numerous consequences include the singularity between the inverse matrices used in estimation methods. Given to this behavior, it is natural to understand that the suitability of these estimators to structural equation models show the same features, either in the simulation results that validate the estimators in different multicollinearity degrees, or in their application to real data. Due to the multicollinearity overview arose from the fact that the matrices inversion is impracticable, the usage of numerical procedures demanded by the maximum likelihood methods leads to numerical singularity problems. An alternative could be the use of the Partial Least Squares (PLS) method, however, it is demanded that the observed variables are built by assuming a positive correlation with the latent variable. Thus, theoretically, it is expected that the load signals are positive, however, there are no restrictions to these signals in the algorithms used in the PLS method. This fact implies in corrective areas, such as the observed variables removal or new formulations of the theoretical model. In view of this problem, this paper aimed to propose adaptations of six generalized ridge estimators as alternative methods to estimate SEM parameters. The conclusion is that the evaluated estimators presented the same performance in terms of accuracy, precision while considering the scenarios represented by model without specification error and model with specification error, different levels of multicollinearity and sample sizes.Multicollinearity is detected via regression models, where independent variables are strongly correlated. Since they entail linear relations between observed or latent variables, the structural equation models (SEM) are subject to the multicollinearity effect, whose numerous consequences include the singularity between the inverse matrices used in estimation methods. Given to this behavior, it is natural to understand that the suitability of these estimators to structural equation models show the same features, either in the simulation results that validate the estimators in different multicollinearity degrees, or in their application to real data. Due to the multicollinearity overview arose from the fact that the matrices inversion is impracticable, the usage of numerical procedures demanded by the maximum likelihood methods leads to numerical singularity problems. An alternative could be the use of the Partial Least Squares (PLS) method, however, it is demanded that the observed variables are built by assuming a positive correlation with the latent variable. Thus, theoretically, it is expected that the load signals are positive, however, there are no restrictions to these signals in the algorithms used in the PLS method. This fact implies in corrective areas, such as the observed variables removal or new formulations of the theoretical model. In view of this problem, this paper aimed to propose adaptations of six generalized ridge estimators as alternative methods to estimate SEM parameters. The conclusion is that the evaluated estimators presented the same performance in terms of accuracy, precision while considering the scenarios represented by model without specification error and model with specification error, different levels of multicollinearity and sample sizes.Universidade Estadual De Maringá2020-08-20info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttp://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/4992910.4025/actascitechnol.v43i1.49929Acta Scientiarum. Technology; Vol 43 (2021): Publicação contínua; e49929Acta Scientiarum. Technology; v. 43 (2021): Publicação contínua; e499291806-25631807-8664reponame:Acta scientiarum. Technology (Online)instname:Universidade Estadual de Maringá (UEM)instacron:UEMenghttp://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/49929/751375150565Copyright (c) 2021 Acta Scientiarum. Technologyhttp://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessPereira, Gislene Araujo Resende, MarianaCirillo, Marcelo Ângelo 2020-10-09T12:48:03Zoai:periodicos.uem.br/ojs:article/49929Revistahttps://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/indexPUBhttps://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/oai||actatech@uem.br1807-86641806-2563opendoar:2020-10-09T12:48:03Acta scientiarum. Technology (Online) - Universidade Estadual de Maringá (UEM)false
dc.title.none.fl_str_mv Generalized ridge estimators adapted in structural equation models
Generalized ridge estimators adapted in structural equation models
title Generalized ridge estimators adapted in structural equation models
spellingShingle Generalized ridge estimators adapted in structural equation models
Pereira, Gislene Araujo
structural model; generalized ridge regression; multicollinearity.
structural model; generalized ridge regression; multicollinearity.
title_short Generalized ridge estimators adapted in structural equation models
title_full Generalized ridge estimators adapted in structural equation models
title_fullStr Generalized ridge estimators adapted in structural equation models
title_full_unstemmed Generalized ridge estimators adapted in structural equation models
title_sort Generalized ridge estimators adapted in structural equation models
author Pereira, Gislene Araujo
author_facet Pereira, Gislene Araujo
Resende, Mariana
Cirillo, Marcelo Ângelo
author_role author
author2 Resende, Mariana
Cirillo, Marcelo Ângelo
author2_role author
author
dc.contributor.author.fl_str_mv Pereira, Gislene Araujo
Resende, Mariana
Cirillo, Marcelo Ângelo
dc.subject.por.fl_str_mv structural model; generalized ridge regression; multicollinearity.
structural model; generalized ridge regression; multicollinearity.
topic structural model; generalized ridge regression; multicollinearity.
structural model; generalized ridge regression; multicollinearity.
description Multicollinearity is detected via regression models, where independent variables are strongly correlated. Since they entail linear relations between observed or latent variables, the structural equation models (SEM) are subject to the multicollinearity effect, whose numerous consequences include the singularity between the inverse matrices used in estimation methods. Given to this behavior, it is natural to understand that the suitability of these estimators to structural equation models show the same features, either in the simulation results that validate the estimators in different multicollinearity degrees, or in their application to real data. Due to the multicollinearity overview arose from the fact that the matrices inversion is impracticable, the usage of numerical procedures demanded by the maximum likelihood methods leads to numerical singularity problems. An alternative could be the use of the Partial Least Squares (PLS) method, however, it is demanded that the observed variables are built by assuming a positive correlation with the latent variable. Thus, theoretically, it is expected that the load signals are positive, however, there are no restrictions to these signals in the algorithms used in the PLS method. This fact implies in corrective areas, such as the observed variables removal or new formulations of the theoretical model. In view of this problem, this paper aimed to propose adaptations of six generalized ridge estimators as alternative methods to estimate SEM parameters. The conclusion is that the evaluated estimators presented the same performance in terms of accuracy, precision while considering the scenarios represented by model without specification error and model with specification error, different levels of multicollinearity and sample sizes.
publishDate 2020
dc.date.none.fl_str_mv 2020-08-20
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/49929
10.4025/actascitechnol.v43i1.49929
url http://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/49929
identifier_str_mv 10.4025/actascitechnol.v43i1.49929
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv http://www.periodicos.uem.br/ojs/index.php/ActaSciTechnol/article/view/49929/751375150565
dc.rights.driver.fl_str_mv Copyright (c) 2021 Acta Scientiarum. Technology
http://creativecommons.org/licenses/by/4.0
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Copyright (c) 2021 Acta Scientiarum. Technology
http://creativecommons.org/licenses/by/4.0
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidade Estadual De Maringá
publisher.none.fl_str_mv Universidade Estadual De Maringá
dc.source.none.fl_str_mv Acta Scientiarum. Technology; Vol 43 (2021): Publicação contínua; e49929
Acta Scientiarum. Technology; v. 43 (2021): Publicação contínua; e49929
1806-2563
1807-8664
reponame:Acta scientiarum. Technology (Online)
instname:Universidade Estadual de Maringá (UEM)
instacron:UEM
instname_str Universidade Estadual de Maringá (UEM)
instacron_str UEM
institution UEM
reponame_str Acta scientiarum. Technology (Online)
collection Acta scientiarum. Technology (Online)
repository.name.fl_str_mv Acta scientiarum. Technology (Online) - Universidade Estadual de Maringá (UEM)
repository.mail.fl_str_mv ||actatech@uem.br
_version_ 1799315337379840000