Alternative smoothing strategies in smooth partial least squares path modelling

Detalhes bibliográficos
Autor(a) principal: Lopes, Tiago Guia Ribeiro
Data de Publicação: 2020
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10362/99737
Resumo: Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Marketing Research e CRM
id RCAP_9ec2899fecaaeacb235385ba4b3d4494
oai_identifier_str oai:run.unl.pt:10362/99737
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Alternative smoothing strategies in smooth partial least squares path modellingPLS-PMNonlinearPLSsSmoothingMonte-Carlo SimulationNatural Cubic SplinesP-SplinesThin Plate Regression SplinesDissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Marketing Research e CRMThe assessment of nonlinear relationships in the context of Partial Least Squares Path Modelling (PLS-PM) has received a growing interest in recent years. One important contribution to this subject has been the work of Henseler, Fassot, Dijkstra and Wilson (2012) on the analysis of four different approaches to quadratic effects. The Smooth Partial Least Squares (PLSs) estimation technique studied in this work removes any assumptions on the structure of the nonlinear relationships between latent variables, by applying smoothing spline techniques to the structural model. Performance results of the PLSs show that it is a powerful tool in the context of predictive research, for instance to support the definition of targeted policies. Building from the hybrid approach to the PLS algorithm introduced by Wold (1982), we compare the performance of alternative spline designs, including natural cubic splines, P-Splines and Thin Plate Regression Splines (TPRS). For this purpose, Monte-Carlo simulations are carried with a conceptual model drawn from a comprehensive set of nonlinear relationships, in different sample sizes. All model configurations are compared using Root Mean Squared Error (RMSE) and absolute bias results. The benchmarking exercise shows that, in most contexts, P-Splines perform slightly better than TPRS and natural cubic splines.Mendes, Jorge MoraisRUNLopes, Tiago Guia Ribeiro2020-06-22T09:33:23Z2020-06-022020-06-02T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10362/99737TID:202487199enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2024-03-11T04:46:30Zoai:run.unl.pt:10362/99737Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T03:39:14.263079Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Alternative smoothing strategies in smooth partial least squares path modelling
title Alternative smoothing strategies in smooth partial least squares path modelling
spellingShingle Alternative smoothing strategies in smooth partial least squares path modelling
Lopes, Tiago Guia Ribeiro
PLS-PM
Nonlinear
PLSs
Smoothing
Monte-Carlo Simulation
Natural Cubic Splines
P-Splines
Thin Plate Regression Splines
title_short Alternative smoothing strategies in smooth partial least squares path modelling
title_full Alternative smoothing strategies in smooth partial least squares path modelling
title_fullStr Alternative smoothing strategies in smooth partial least squares path modelling
title_full_unstemmed Alternative smoothing strategies in smooth partial least squares path modelling
title_sort Alternative smoothing strategies in smooth partial least squares path modelling
author Lopes, Tiago Guia Ribeiro
author_facet Lopes, Tiago Guia Ribeiro
author_role author
dc.contributor.none.fl_str_mv Mendes, Jorge Morais
RUN
dc.contributor.author.fl_str_mv Lopes, Tiago Guia Ribeiro
dc.subject.por.fl_str_mv PLS-PM
Nonlinear
PLSs
Smoothing
Monte-Carlo Simulation
Natural Cubic Splines
P-Splines
Thin Plate Regression Splines
topic PLS-PM
Nonlinear
PLSs
Smoothing
Monte-Carlo Simulation
Natural Cubic Splines
P-Splines
Thin Plate Regression Splines
description Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Marketing Research e CRM
publishDate 2020
dc.date.none.fl_str_mv 2020-06-22T09:33:23Z
2020-06-02
2020-06-02T00:00:00Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10362/99737
TID:202487199
url http://hdl.handle.net/10362/99737
identifier_str_mv TID:202487199
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799138008424775680

Registros relacionados