Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms
Using a comparative experimental design, four neural network architectures—feedforward, LSTM, sequential feedforward, and robust LSTM—are evaluated to predict the provincial energy intensity index, based on panel data from 2018 to 2023 and key variables related to energy consumption, economic activi...
Zapisane w:
| 1. autor: | |
|---|---|
| Kolejni autorzy: | |
| Format: | article |
| Język: | spa |
| Wydane: |
2025
|
| Hasła przedmiotowe: | |
| Dostęp online: | https://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/502 |
| Etykiety: |
Dodaj etykietę
Nie ma etykietki, Dołącz pierwszą etykiete!
|
Dodaj etykietę | _version_ | 1858113931501895680 |
|---|---|
| author | Rodríguez, Lester |
| author2 | Litardo Unda, Julio Goivanni |
| author2_role | author |
| author_facet | Rodríguez, Lester Litardo Unda, Julio Goivanni |
| author_role | author |
| collection | Revista Cuestiones Económicas |
| dc.creator.none.fl_str_mv | Rodríguez, Lester Litardo Unda, Julio Goivanni |
| dc.date.none.fl_str_mv | 2025-06-30 |
| dc.format.none.fl_str_mv | application/pdf |
| dc.identifier.none.fl_str_mv | https://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/502 10.47550/35.1.8 |
| dc.language.none.fl_str_mv | spa |
| dc.publisher.none.fl_str_mv | Banco Central del Ecuador |
| dc.relation.none.fl_str_mv | https://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/502/392 |
| dc.rights.none.fl_str_mv | Derechos de autor 2025 Cuestiones Económicas https://creativecommons.org/licenses/by-nc/4.0 info:eu-repo/semantics/openAccess |
| dc.source.none.fl_str_mv | Cuestiones Económicas; Vol. 35 Núm. 1 (2025): Revista Cuestiones Económicas; Autores: Xavier Rodriguez-Cruz y Julio Litardo 2697-3367 2697-3367 reponame:Revista Cuestiones Económicas instname:Banco Central del Ecuador instacron:BCE |
| dc.subject.none.fl_str_mv | Energy Intensity Artificial Neural Networks Machine Learning K-means DBSCAN Intensidad energética Redes neuronales artificiales Aprendizaje automático K-means DBSCAN |
| dc.title.none.fl_str_mv | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms Modelado y Predicción del Índice de Intensidad Energética por Provincias del Ecuador Utilizando Redes Neuronales y Análisis de Conglomerados con Algoritmos de Aprendizaje Automático |
| dc.type.none.fl_str_mv | info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Artículos de Investigación |
| description | Using a comparative experimental design, four neural network architectures—feedforward, LSTM, sequential feedforward, and robust LSTM—are evaluated to predict the provincial energy intensity index, based on panel data from 2018 to 2023 and key variables related to energy consumption, economic activity, and demographic dynamics. The sequential feedforward model demonstrated the highest predictive accuracy (R2 = 0.724), outperforming recurrent approaches in capturing temporal patterns, precision, and computational efficiency. Clustering methods allow for the characterization of provinces’ behavior in terms of energy intensity, identifying groups with similar energy con sumption patterns relative to their economic output. Orellana, Pastaza, Santa Elena, and Zamora Chinchipe stand out for exhibiting distinct or atypical energy profiles, highlighting the usefulness of algorithms such as DBSCAN in detecting unique regional dynamics.The results help identify regional differences in energy efficiency, providing an empirical basis for designing and implementing tailored energy policies. These insights are especially useful in varied geographic and socio-economic settings, where encouraging sustainable energy use and reducing disparities in distribution and consumption are key objectives. |
| eu_rights_str_mv | openAccess |
| format | article |
| id | REVCUESTEC_5b3a03dc218945c18bd58551cdba0827 |
| identifier_str_mv | 10.47550/35.1.8 |
| instacron_str | BCE |
| institution | BCE |
| instname_str | Banco Central del Ecuador |
| language | spa |
| network_acronym_str | REVCUESTEC |
| network_name_str | Revista Cuestiones Económicas |
| oai_identifier_str | oai:estudioseconomicos.bce.fin.ec:article/502 |
| publishDate | 2025 |
| publisher.none.fl_str_mv | Banco Central del Ecuador |
| reponame_str | Revista Cuestiones Económicas |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | Revista Cuestiones Económicas - Banco Central del Ecuador |
| repository_id_str | |
| rights_invalid_str_mv | Derechos de autor 2025 Cuestiones Económicas https://creativecommons.org/licenses/by-nc/4.0 |
| spelling | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning AlgorithmsModelado y Predicción del Índice de Intensidad Energética por Provincias del Ecuador Utilizando Redes Neuronales y Análisis de Conglomerados con Algoritmos de Aprendizaje AutomáticoRodríguez, LesterLitardo Unda, Julio GoivanniEnergy IntensityArtificial Neural NetworksMachine LearningK-meansDBSCANIntensidad energéticaRedes neuronales artificialesAprendizaje automáticoK-meansDBSCANUsing a comparative experimental design, four neural network architectures—feedforward, LSTM, sequential feedforward, and robust LSTM—are evaluated to predict the provincial energy intensity index, based on panel data from 2018 to 2023 and key variables related to energy consumption, economic activity, and demographic dynamics. The sequential feedforward model demonstrated the highest predictive accuracy (R2 = 0.724), outperforming recurrent approaches in capturing temporal patterns, precision, and computational efficiency. Clustering methods allow for the characterization of provinces’ behavior in terms of energy intensity, identifying groups with similar energy con sumption patterns relative to their economic output. Orellana, Pastaza, Santa Elena, and Zamora Chinchipe stand out for exhibiting distinct or atypical energy profiles, highlighting the usefulness of algorithms such as DBSCAN in detecting unique regional dynamics.The results help identify regional differences in energy efficiency, providing an empirical basis for designing and implementing tailored energy policies. These insights are especially useful in varied geographic and socio-economic settings, where encouraging sustainable energy use and reducing disparities in distribution and consumption are key objectives.Mediante un diseño experimental comparativo, se evalúan cuatro arquitecturas de redes neuronales feedforward, LSTM, feedforward secuencial y LSTM robusto para predecir el índice de intensidad energética provincial, utilizando datos de panel de 2018 a 2023 con variables claves relacionadas con el consumo energético, la actividad económica y la dinámica demográfica. El modelo feedforward secuencial mostró mayor precisión predictiva (R2 = 0,724), superando el rendimiento de los enfoques recurrentes en la capacidad para capturar patrones temporales, precisión y eficiencia computacional. Los métodos de agrupamiento permiten caracterizar el comportamiento de las provincias en términos de intensidad energética, identificando grupos con patrones de consumo energético similares en relación con su producción económica. Orellana, Pastaza, Santa Elena y Zamora Chinchipe destacan por presentar perfiles energéticos diferenciados o atípicos, resaltando la utilidad de algoritmos como DBSCAN para detectar dinámicas regionales singulares.Los resultados permiten identificar asimetrías territoriales en eficiencia energética, proporcionando una base empírica para el diseño e implementación de políticas energéticas diferenciadas. Estos conocimientos son particularmente valiosos en contextos geográficos y socioeconómicos diversos, donde promover el uso sostenible de la energía y reducir las desigualdades en la distribución y el consumo son objetivos fundamentales.Banco Central del Ecuador2025-06-30info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionArtículos de Investigaciónapplication/pdfhttps://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/50210.47550/35.1.8Cuestiones Económicas; Vol. 35 Núm. 1 (2025): Revista Cuestiones Económicas; Autores: Xavier Rodriguez-Cruz y Julio Litardo2697-33672697-3367reponame:Revista Cuestiones Económicasinstname:Banco Central del Ecuadorinstacron:BCEspahttps://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/502/392Derechos de autor 2025 Cuestiones Económicashttps://creativecommons.org/licenses/by-nc/4.0info:eu-repo/semantics/openAccess2025-09-02T16:01:32Zoai:estudioseconomicos.bce.fin.ec:article/502Portal de revistashttps://estudioseconomicos.bce.fin.ec/index.php/RevistaCEOrganismo de gobiernowww.bce.fin.echttps://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/oaiEcuadoropendoar:2025-09-02T16:01:32falsePortal de revistashttps://estudioseconomicos.bce.fin.ec/index.php/RevistaCEOrganismo de gobiernowww.bce.fin.echttps://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/oaiEcuadoropendoar:2025-09-02T16:01:32Revista Cuestiones Económicas - Banco Central del Ecuadorfalse |
| spellingShingle | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms Rodríguez, Lester Energy Intensity Artificial Neural Networks Machine Learning K-means DBSCAN Intensidad energética Redes neuronales artificiales Aprendizaje automático K-means DBSCAN |
| status_str | publishedVersion |
| title | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms |
| title_full | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms |
| title_fullStr | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms |
| title_full_unstemmed | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms |
| title_short | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms |
| title_sort | Modeling and Prediction of the Energy Intensity Index by Provinces of Ecuador Using Neural Networks and Cluster Analysis with Machine Learning Algorithms |
| topic | Energy Intensity Artificial Neural Networks Machine Learning K-means DBSCAN Intensidad energética Redes neuronales artificiales Aprendizaje automático K-means DBSCAN |
| url | https://estudioseconomicos.bce.fin.ec/index.php/RevistaCE/article/view/502 |