Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz

Detalhes bibliográficos
Autor(a) principal: Azevedo, Maria Margarida Marques Campos de
Data de Publicação: 2021
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10451/49336
Resumo: Tese de mestrado, Bioestatística, Universidade de Lisboa, Faculdade de Ciências, 2021
id RCAP_aef71b0f129bb72b7d56b5429a264b7a
oai_identifier_str oai:repositorio.ul.pt:10451/49336
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda OrdazGLMGAMAranda-OrdazCzadoTrabalhos de projeto de mestrado - 2021Departamento de Estatística e Investigação OperacionalTese de mestrado, Bioestatística, Universidade de Lisboa, Faculdade de Ciências, 2021For a good clinical practice and management to be attained, statistical analysis can perform a major role. Statistical models can greatly aid in medical decision making, as for example in the admission of critically ill patients to the Intensive Care Units (ICUs). These services have the mission of providing health care to patients in critical situations, which constitutes a challenge to hospital management, considering the heavy budgets that are necessary to maintain the quality of response. Therefore, decisions are made daily bearing in mind the effectiveness of the treatment versus its cost. In order to help the decision-making process metrics can be obtained, usually via Generalized Linear Models (GLMs) and Generalized Additive Models (GAMs). These are generally oriented towards the quantification of the risk of mortality and are characterized by a small number of variables, from which a score is extracted that reflects the patient's state of severity in addition to an in-hospital mortality estimate. Among the components of GLMs and GAMs which can be focused on in order to improve the quality of the models, the link function is highlighted. Recent work using models with flexible parametric link functions, namely with link functions belonging to the family of asymmetric functions of Aranda-Ordaz, showed an improvement in their performance. On the other hand, studies involving link functions belonging to the family of Czado functions are scarce. Using a Czado link function provides a greater flexibility, by it depending on three parameters, rather than the Aranda-Ordaz link function, which merely depends on one. Thus, a comparative analysis of the performance of both models referred (GLM, GAM), using the Aranda-Ordaz and Czado link functions, and considering the Logistic link function as a baseline was the primary line of work. The results presented themselves as inconclusive regarding the greater performance of either of the link functions, which can be related to the data used and not necessarily the actual performance of the models and link functions. Further studies should be carried using different data sets in order to truly access the performance of the models using both Czado and Aranda-Ordaz link functions.Sousa, Lisete Maria Ribeiro de, 1972-Geraldes, CarlosRepositório da Universidade de LisboaAzevedo, Maria Margarida Marques Campos de2021-08-26T09:47:58Z202120212021-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10451/49336TID:202931153enginfo: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:RCAAP2023-11-08T16:53:05Zoai:repositorio.ul.pt:10451/49336Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T22:01:00.356391Repositório 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 Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
title Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
spellingShingle Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
Azevedo, Maria Margarida Marques Campos de
GLM
GAM
Aranda-Ordaz
Czado
Trabalhos de projeto de mestrado - 2021
Departamento de Estatística e Investigação Operacional
title_short Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
title_full Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
title_fullStr Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
title_full_unstemmed Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
title_sort Comparative analysis of non-linear models with parametric link function based on the family of Czado functions and asymmetric functions of Aranda Ordaz
author Azevedo, Maria Margarida Marques Campos de
author_facet Azevedo, Maria Margarida Marques Campos de
author_role author
dc.contributor.none.fl_str_mv Sousa, Lisete Maria Ribeiro de, 1972-
Geraldes, Carlos
Repositório da Universidade de Lisboa
dc.contributor.author.fl_str_mv Azevedo, Maria Margarida Marques Campos de
dc.subject.por.fl_str_mv GLM
GAM
Aranda-Ordaz
Czado
Trabalhos de projeto de mestrado - 2021
Departamento de Estatística e Investigação Operacional
topic GLM
GAM
Aranda-Ordaz
Czado
Trabalhos de projeto de mestrado - 2021
Departamento de Estatística e Investigação Operacional
description Tese de mestrado, Bioestatística, Universidade de Lisboa, Faculdade de Ciências, 2021
publishDate 2021
dc.date.none.fl_str_mv 2021-08-26T09:47:58Z
2021
2021
2021-01-01T00:00:00Z
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/10451/49336
TID:202931153
url http://hdl.handle.net/10451/49336
identifier_str_mv TID:202931153
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_ 1799134558018338816

Registros relacionados