Data pre-processing for neural network-based forecasting: does it really matter?

This study aims to analyze the effects of data pre-processing on the forecasting performance of neural network models. We use three different Artificial Neural Networks techniques to predict tourist demand: multi-layer perceptron, radial basis function and Elman neural networks. The structure of the...

Descripción completa

Detalles Bibliográficos
Autores: Clavería González, Óscar, Monte Moreno, Enric, Torra Porras, Salvador
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2017
País:España
Institución:Universidad de Barcelona
Repositorio:Dipòsit Digital de la UB
OAI Identifier:oai:diposit.ub.edu:2445/96752
Acceso en línea:https://hdl.handle.net/2445/96752
Access Level:acceso abierto
Palabra clave:Previsió econòmica
Desenvolupament econòmic
Xarxes neuronals (Informàtica)
Economic forecasting
Economic development
Neural networks (Computer science)
id ES_bcb9d44a0a769dea41ed2d93c43b186b
oai_identifier_str oai:diposit.ub.edu:2445/96752
network_acronym_str ES
network_name_str España
repository_id_str
spelling Data pre-processing for neural network-based forecasting: does it really matter?Clavería González, ÓscarMonte Moreno, EnricTorra Porras, SalvadorPrevisió econòmicaDesenvolupament econòmicXarxes neuronals (Informàtica)Economic forecastingEconomic developmentNeural networks (Computer science)This study aims to analyze the effects of data pre-processing on the forecasting performance of neural network models. We use three different Artificial Neural Networks techniques to predict tourist demand: multi-layer perceptron, radial basis function and Elman neural networks. The structure of the networks is based on a multiple-output approach. We use official statistical data of inbound international tourism demand to Catalonia (Spain) and compare the forecasting accuracy of four processing methods for the input vector of the networks: levels, growth rates, seasonally adjusted levels and seasonally adjusted growth rates. When comparing the forecasting accuracy of the different inputs for each visitor market and for different forecasting horizons, we obtain significantly better forecasts with levels than with growth rates. We also find that seasonally adjusted series significantly improve the forecasting performance of the networks, which hints at the significance of deseasonalizing the time series when using neural networks with forecasting purposes. These results reveal that, when using seasonal data, neural networks performance can be significantly improved by working directly with seasonally adjusted levels.Taylor and Francis2017info:eu-repo/semantics/articleinfo:eu-repo/semantics/acceptedVersionapplication/pdfhttps://hdl.handle.net/2445/96752Articles publicats en revistes (Econometria, Estadística i Economia Aplicada)reponame:Dipòsit Digital de la UBinstname:Universidad de BarcelonaInglésVersió postprint del document publicat a: http://www.tandfonline.com/doi/abs/10.3846/20294913.2015.1070772Technological and Economic Development of Economy, 2017, vol. 23, núm. 5, p. 709-725http://dx.doi.org/10.3846/20294913.2015.1070772(c) Vilnius Gediminas Technical University, 2017info:eu-repo/semantics/openAccessoai:diposit.ub.edu:2445/967522026-05-27T06:46:51Z
dc.title.none.fl_str_mv Data pre-processing for neural network-based forecasting: does it really matter?
title Data pre-processing for neural network-based forecasting: does it really matter?
spellingShingle Data pre-processing for neural network-based forecasting: does it really matter?
Clavería González, Óscar
Previsió econòmica
Desenvolupament econòmic
Xarxes neuronals (Informàtica)
Economic forecasting
Economic development
Neural networks (Computer science)
title_short Data pre-processing for neural network-based forecasting: does it really matter?
title_full Data pre-processing for neural network-based forecasting: does it really matter?
title_fullStr Data pre-processing for neural network-based forecasting: does it really matter?
title_full_unstemmed Data pre-processing for neural network-based forecasting: does it really matter?
title_sort Data pre-processing for neural network-based forecasting: does it really matter?
dc.creator.none.fl_str_mv Clavería González, Óscar
Monte Moreno, Enric
Torra Porras, Salvador
author Clavería González, Óscar
author_facet Clavería González, Óscar
Monte Moreno, Enric
Torra Porras, Salvador
author_role author
author2 Monte Moreno, Enric
Torra Porras, Salvador
author2_role author
author
dc.subject.none.fl_str_mv Previsió econòmica
Desenvolupament econòmic
Xarxes neuronals (Informàtica)
Economic forecasting
Economic development
Neural networks (Computer science)
topic Previsió econòmica
Desenvolupament econòmic
Xarxes neuronals (Informàtica)
Economic forecasting
Economic development
Neural networks (Computer science)
description This study aims to analyze the effects of data pre-processing on the forecasting performance of neural network models. We use three different Artificial Neural Networks techniques to predict tourist demand: multi-layer perceptron, radial basis function and Elman neural networks. The structure of the networks is based on a multiple-output approach. We use official statistical data of inbound international tourism demand to Catalonia (Spain) and compare the forecasting accuracy of four processing methods for the input vector of the networks: levels, growth rates, seasonally adjusted levels and seasonally adjusted growth rates. When comparing the forecasting accuracy of the different inputs for each visitor market and for different forecasting horizons, we obtain significantly better forecasts with levels than with growth rates. We also find that seasonally adjusted series significantly improve the forecasting performance of the networks, which hints at the significance of deseasonalizing the time series when using neural networks with forecasting purposes. These results reveal that, when using seasonal data, neural networks performance can be significantly improved by working directly with seasonally adjusted levels.
publishDate 2017
dc.date.none.fl_str_mv 2017
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/acceptedVersion
format article
status_str acceptedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/2445/96752
url https://hdl.handle.net/2445/96752
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Versió postprint del document publicat a: http://www.tandfonline.com/doi/abs/10.3846/20294913.2015.1070772
Technological and Economic Development of Economy, 2017, vol. 23, núm. 5, p. 709-725
http://dx.doi.org/10.3846/20294913.2015.1070772
dc.rights.none.fl_str_mv (c) Vilnius Gediminas Technical University, 2017
info:eu-repo/semantics/openAccess
rights_invalid_str_mv (c) Vilnius Gediminas Technical University, 2017
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Taylor and Francis
publisher.none.fl_str_mv Taylor and Francis
dc.source.none.fl_str_mv Articles publicats en revistes (Econometria, Estadística i Economia Aplicada)
reponame:Dipòsit Digital de la UB
instname:Universidad de Barcelona
instname_str Universidad de Barcelona
reponame_str Dipòsit Digital de la UB
collection Dipòsit Digital de la UB
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869418138722893824
score 15,198674

Similares en LA Referencia