Estimation of a constrained multinomial logit model

Identifying the set of available alternatives in a choice process after considering an individual's bounds or thresholds is a complex process that, in practice, is commonly simplified by assuming exogenous rules in the choice set formation. The Constrained Multinomial Logit (CMNL) model incorpo...

Descripción completa

Detalles Bibliográficos
Autores: Castro, Marisol, Martinez-Concha, Francisco Javier, Munizaga-Muñoz, Marcela Adriana
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2013
País:Chile
Idioma:inglés
OAI Identifier:oai:repositorio.anid.cl:10533/197068
Acceso en línea:https://hdl.handle.net/10533/197068
Access Level:acceso abierto
id CL_ea3670e698939cd39f5373a1cb7fe264
oai_identifier_str oai:repositorio.anid.cl:10533/197068
network_acronym_str CL
network_name_str Chile
repository_id_str
spelling Munizaga-Muñoz, Marcela AdrianaMartinez-Concha, Francisco JavierCastro, Marisol201310.1007/s11116-012-9435-4https://hdl.handle.net/10533/197068http://purl.org/coar/access_right/c_abf2Estimation of a constrained multinomial logit model581563NLDNEW YORKCastro, MarisolMartinez-Concha, Francisco JavierMunizaga-Muñoz, Marcela Adriana2017-04-27T18:50:15Z2022-07-07T01:44:30Z2017-04-27T18:50:15Z2022-07-07T01:44:30Z2013Identifying the set of available alternatives in a choice process after considering an individual's bounds or thresholds is a complex process that, in practice, is commonly simplified by assuming exogenous rules in the choice set formation. The Constrained Multinomial Logit (CMNL) model incorporates thresholds in several attributes as a key endogenous process to define the alternatives choice/rejection mechanism. The model allows for the inclusion of multiple constraints and has a closed form. In this paper, we study the estimation of the CMNL model using the maximum likelihood function, develop a methodology to estimate the model overcoming identification problems by an endogenous partition of the sample, and test the model estimation with both synthetic and real data. The CMNL model appears to be suitable for general applications as it presents a significantly better fit than the MNL model under constrained behaviour and replicates the MNL estimates in the unconstrained case. Using mode choice real data, we found significant differences in the values of times and elasticities between compensatory MNL and semi-compensatory CMNL models, which increase as the thresholds on attributes become active.Funding: Fondecyt (1110124, 1120288), ISCI (ICM P-05-004-F, CONICYT FBO16), FONDEF D10I-1002. A previous version of this paper was presented in the International Choice Modelling Conference. The authors wish to thank the comments of anonymous referees.3FONDEFmarisolcastro@gmail.com; fmartine@ing.uchile.cl; mamuniza@ing.uchile.clFondecyt [1110124, 1120288]; ISCI [ICM P-05-004-F, CONICYT FBO16]; FONDEF [D10I-1002]FONDEF0D10I1002D10I1002virtual::42143-1WOS:000317750600005WOS:0003177506000050049-4488https://hdl.handle.net/10533/197068SPRINGERinstname: Conicytreponame: Repositorio Digital RI2.0instname: Conicytreponame: Repositorio Digital RI2.010.1007/s11116-012-9435-4info:eu-repo/grantAgreement/Fondef/D10I1002info:eu-repo/semantics/dataset/hdl.handle.net/10533/93477https://doi.org/10.1007/s11116-012-9435-4info:eu-repo/semantics/openAccessEstimation of a constrained multinomial logit modelTRANSPORTATIONTransportationArticuloinfo:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionengArticulohttps://hdl.handle.net/10533/197068FONDEFhttp://purl.org/coar/resource_type/c_2df8fbb1098f96d1-3e02-49e2-9ac4-3f4fc955cd49virtual::42143-1098f96d1-3e02-49e2-9ac4-3f4fc955cd49virtual::42143-110533/197068oai:repositorio.anid.cl:10533/1970682023-07-24 16:55:32.969https://repositorio.anid.clRepositorio ANIDaletelier@anid.cl
dc.title.none.fl_str_mv Estimation of a constrained multinomial logit model
dc.title.journal.none.fl_str_mv TRANSPORTATION
dc.title.journalabbreviation.none.fl_str_mv Transportation
title Estimation of a constrained multinomial logit model
spellingShingle Estimation of a constrained multinomial logit model
Castro, Marisol
title_short Estimation of a constrained multinomial logit model
title_full Estimation of a constrained multinomial logit model
title_fullStr Estimation of a constrained multinomial logit model
title_full_unstemmed Estimation of a constrained multinomial logit model
title_sort Estimation of a constrained multinomial logit model
dc.creator.none.fl_str_mv Castro, Marisol
Martinez-Concha, Francisco Javier
Munizaga-Muñoz, Marcela Adriana
author Castro, Marisol
author_facet Castro, Marisol
Martinez-Concha, Francisco Javier
Munizaga-Muñoz, Marcela Adriana
author_role author
author2 Martinez-Concha, Francisco Javier
Munizaga-Muñoz, Marcela Adriana
author2_role author
author
description Identifying the set of available alternatives in a choice process after considering an individual's bounds or thresholds is a complex process that, in practice, is commonly simplified by assuming exogenous rules in the choice set formation. The Constrained Multinomial Logit (CMNL) model incorporates thresholds in several attributes as a key endogenous process to define the alternatives choice/rejection mechanism. The model allows for the inclusion of multiple constraints and has a closed form. In this paper, we study the estimation of the CMNL model using the maximum likelihood function, develop a methodology to estimate the model overcoming identification problems by an endogenous partition of the sample, and test the model estimation with both synthetic and real data. The CMNL model appears to be suitable for general applications as it presents a significantly better fit than the MNL model under constrained behaviour and replicates the MNL estimates in the unconstrained case. Using mode choice real data, we found significant differences in the values of times and elasticities between compensatory MNL and semi-compensatory CMNL models, which increase as the thresholds on attributes become active.
publishDate 2013
dc.date.issued.none.fl_str_mv 2013
dc.date.accessioned.none.fl_str_mv 2017-04-27T18:50:15Z
2022-07-07T01:44:30Z
dc.date.available.none.fl_str_mv 2017-04-27T18:50:15Z
2022-07-07T01:44:30Z
dc.type.none.fl_str_mv Articulo
dc.type.driver.none.fl_str_mv info:eu-repo/semantics/article
dc.type.openaire.none.fl_str_mv info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.eissn.none.fl_str_mv 0
dc.identifier.folio.none.fl_str_mv D10I1002
D10I1002
dc.identifier.idwos.none.fl_str_mv WOS:000317750600005
dc.identifier.isi.none.fl_str_mv WOS:000317750600005
dc.identifier.issn.none.fl_str_mv 0049-4488
dc.identifier.uri.none.fl_str_mv https://hdl.handle.net/10533/197068
identifier_str_mv 0
D10I1002
WOS:000317750600005
0049-4488
url https://hdl.handle.net/10533/197068
dc.language.*.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv instname: Conicyt
reponame: Repositorio Digital RI2.0
instname: Conicyt
reponame: Repositorio Digital RI2.0
dc.relation.doi.none.fl_str_mv 10.1007/s11116-012-9435-4
dc.relation.projectid.none.fl_str_mv info:eu-repo/grantAgreement/Fondef/D10I1002
dc.relation.set.none.fl_str_mv info:eu-repo/semantics/dataset/hdl.handle.net/10533/93477
dc.relation.uri.none.fl_str_mv https://doi.org/10.1007/s11116-012-9435-4
dc.rights.driver.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.coverage.spatial.none.fl_str_mv NEW YORK
dc.publisher.none.fl_str_mv SPRINGER
publisher.none.fl_str_mv SPRINGER
repository.name.fl_str_mv Repositorio ANID
repository.mail.fl_str_mv aletelier@anid.cl
_version_ 1843518734029815808
score 15,811543

Similares en LA Referencia