eprintid: 9236 rev_number: 8 eprint_status: archive userid: 2 dir: disk0/00/00/92/36 datestamp: 2023-10-17 23:30:23 lastmod: 2023-10-17 23:30:24 status_changed: 2023-10-17 23:30:23 type: article metadata_visibility: show creators_name: Joshi, Devendra creators_name: Chithaluru, Premkumar creators_name: Anand, Divya creators_name: Hajjej, Fahima creators_name: Aggarwal, Kapil creators_name: Yélamos Torres, Vanessa creators_name: Bautista Thompson, Ernesto creators_id: creators_id: creators_id: divya.anand@uneatlantico.es creators_id: creators_id: creators_id: vanessa.yelamos@funiber.org creators_id: ernesto.bautista@unini.edu.mx title: An Evolutionary Technique for Building Neural Network Models for Predicting Metal Prices ispublished: pub subjects: uneat_eng divisions: uneatlantico_produccion_cientifica full_text_status: public keywords: evolutionary algorithm; auto-correlation; metal price; neural network; cross-validation; entropy abstract: In this research, a neural network (NN) model for metal price forecasting based on an evolutionary approach is proposed. Both the neural network model’s network parameters and network architecture are selected automatically. The time series metal price data set is used to construct a novel fitness function that takes into account both error minimizations and the reproduction of the auto-correlation function. Calculating the average entropy values allowed the selection of the input parameter count for the neural network model. Gold price forecasting was performed using the proposed methodology. The optimal hidden node number, learning rate, and momentum are 9, 0.026, and 0.76, respectively, according to the evolutionary-based NN model. The proposed strategy is shown to reduce estimation error while also reproducing the auto-correlation function of the time series data set by the validation results with gold price data. The performance of the proposed method is better than other current methods, according to a comparison study. date: 2023-03 publication: Mathematics volume: 11 number: 7 pagerange: 1675 id_number: doi:10.3390/math11071675 refereed: TRUE issn: 2227-7390 official_url: http://doi.org/10.3390/math11071675 access: open language: en citation: Artículo Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica Abierto Inglés In this research, a neural network (NN) model for metal price forecasting based on an evolutionary approach is proposed. Both the neural network model’s network parameters and network architecture are selected automatically. The time series metal price data set is used to construct a novel fitness function that takes into account both error minimizations and the reproduction of the auto-correlation function. Calculating the average entropy values allowed the selection of the input parameter count for the neural network model. Gold price forecasting was performed using the proposed methodology. The optimal hidden node number, learning rate, and momentum are 9, 0.026, and 0.76, respectively, according to the evolutionary-based NN model. The proposed strategy is shown to reduce estimation error while also reproducing the auto-correlation function of the time series data set by the validation results with gold price data. The performance of the proposed method is better than other current methods, according to a comparison study. metadata Joshi, Devendra; Chithaluru, Premkumar; Anand, Divya; Hajjej, Fahima; Aggarwal, Kapil; Yélamos Torres, Vanessa y Bautista Thompson, Ernesto mail SIN ESPECIFICAR, SIN ESPECIFICAR, divya.anand@uneatlantico.es, SIN ESPECIFICAR, SIN ESPECIFICAR, vanessa.yelamos@funiber.org, ernesto.bautista@unini.edu.mx (2023) An Evolutionary Technique for Building Neural Network Models for Predicting Metal Prices. Mathematics, 11 (7). p. 1675. ISSN 2227-7390 document_url: http://repositorio.uneatlantico.es/id/eprint/9236/1/mathematics-11-01675-v2.pdf