Forecast combination using Optimization techniques

dc.contributor.authorValencia-Cárdenas, Marisolspa
dc.contributor.authorCorrea-Morales, Juan Carlosspa
dc.coverage.sucursalUniversidad Nacional de Colombia - Sede Medellínspa
dc.date.accessioned2020-02-07T13:55:53Zspa
dc.date.available2020-02-07T13:55:53Zspa
dc.date.issued2020-01-30spa
dc.description.abstractActualmente existen diversas metodologías de pronóstico, que van desde el conocimiento empírico hasta métodos innovadores, individuales o combinados, que demuestran resultados óptimos. Este documento se deriva de un proceso de investigación y presenta alternativas relacionadas con las combinaciones de pronósticos, utilizando metaheurísticas, por ejemplo, mediante la búsqueda tabú y la programación evolutiva para optimizar el pronóstico. El documento presenta pronósticos combinados basados en la programación evolutiva utilizando mezclas de modelos de regresión bayesiana y modelos de regresión lineal clásico, el modelo de media móvil integrado autorregresivo, el suavizado exponencial y la regresión bayesiana. El documento presenta dos artículos derivados de investigación, la primera compara el algoritmo combinado con los resultados individuales de estos modelos individuales y con la combinación de Bates y Granger utilizando un indicador de error y el valor simétrico de error absoluto medio. Esos modelos y la combinación se aplicaron a la simulación de series temporales y a un caso real de ventas de productos lácteos, generando así pronósticos combinados multiproductos tanto para la simulación como para el caso real. La nueva combinación combinada con la metaheurística evolutiva mostró mejores resultados que los de los otros que se utilizaron. La segunda investigación utiliza series de tiempo simuladas, diseñando dos metaheurísticas basadas en la lista Tabú, que aprenden de los datos con base en el comportamiento estadístico de éstos, como el cluster, así como del mismo valor optimizado del error de ajuste, y se comparan las combinaciones de pronósticos con resultados de modelos individuales a tres tipos de series de tiempo.spa
dc.description.abstractCurrently diverse forecasting methodologies exists, going from the empirical knowledge to the innovative methods, individual or combined, demonstrating optimal results. This document is derived from a research process, and presents alternatives related to forecast combinations, using metaheuristics, for example, by using Tabu search and Evolutive programing to optimize forecasting. One of the designed process consists of creating combination forecasts based on evolutionary programming using, first, a mixture of Bayesian regression models and, second, a mixture of the classical linear regression model, the autoregressive integrated moving average model, exponential smoothing and Bayesian regression. The first research compares the novel combined algorithm with the individual results of these individual models and with the Bates and Granger combination using an error indicator and the symmetrical mean absolute error value. Those models and the novel design were applied to time series simulation and to a real case of dairy products sales, thus generating multiproduct combination forecasts for both the simulation and the real case. The novel combination combined with the evolutionary metaheuristic showed better results than those of the others that were used. The second research uses simulated time series and other metaheuristic that learns from the data an statistical behavior.eng
dc.description.degreelevelDoctoradospa
dc.description.sponsorshipTecnológico de Antioquia, Universidad Nacional de Colombia.spa
dc.format.extent41 páginasspa
dc.format.mimetypeapplication/pdfspa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/75559
dc.language.isoengspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Medellínspa
dc.publisher.departmentEscuela de estadísticaspa
dc.relation.referencesAgarwal, D. K., Silander, J. A., Gelfand, A. E., Dewar, R. E., & Mickelson, J. G. (2005). Tropical deforestation in Madagascar: Analysis using hierarchical, spatially explicit, Bayesian regression models. Ecological Modelling, 185(1), 105–131. http://doi.org/10.1016/j.ecolmodel.2004.11.023spa
dc.relation.referencesAllahverdi, A., & Al-Anzi, F. S. (2006). Evolutionary heuristics and an algorithm for the two-stage assembly scheduling problem to minimize makespan with setup times. International Journal of Production Research, 44(22), 4713–4735. http://doi.org/10.1080/00207540600621029spa
dc.relation.referencesBarrow, D. K., & Kourentzes, N. (2016). Distributions of forecasting errors of forecast combinations: Implications for inventory management. International Journal of Production Economics, 177, 24–33. http://doi.org/10.1016/j.ijpe.2016.03.017spa
dc.relation.referencesBowerman, B., Koehler, A., & O’Connell, R. (2007). Pronósticos, series de tiempo y regresión: un enfoque aplicado. Mexico, DF:. CENCAGE Learning.spa
dc.relation.referencesCang, S., & Yu, H. (2014). A combination selection algorithm on forecasting. European Journal of Operational Research, 234(1), 127–139. http://doi.org/10.1016/j.ejor.2013.08.045spa
dc.relation.referencesChelouah, R., & Siarry, P. (2005). A hybrid method combining continuous tabu search and Nelder– Mead simplex algorithms for the global optimization of multiminima functions. European Journal of Operational Research, 161(3), 636–654. http://doi.org/10.1016/j.ejor.2003.08.053spa
dc.relation.referencesFerragina, A., de los Campos, G., Vazquez, A. I., Cecchinato, A., & Bittante, G. (2015). Bayesian regression models outperform partial least squares methods for predicting milk components and technological properties using infrared spectral data. Journal of Dairy Science, 98(11), 8133–8151. http://doi.org/10.3168/jds.2014-9143spa
dc.relation.referencesGill, J. (2007). Bayesian methods: A social and behavioral sciences approach. United States of America: CHapman & Hall. 2nd. Edspa
dc.relation.referencesHsiao, C., & Wan, S. K. (2014). Is there an optimal forecast combination? Journal of Econometrics, 178(PART 2), 294–309. http://doi.org/10.1016/j.jeconom.2013.11.003spa
dc.relation.referencesKociecki, A., Kolasa, M., & Rubaszek, M. (2012). A Bayesian method of combining judgmental and model-based density forecasts. Economic Modelling, 29, 1349–1355. http://doi.org/10.1016/j.econmod.2012.03.004spa
dc.relation.referencesMelo, L. F., Loaiza, R. A., & Villamizar-Villegas, M. (2016). Bayesian combination for inflation forecasts: The effects of a prior based on central banks’ estimates. Economic Systems. http://doi.org/10.1016/j.ecosys.2015.11.002spa
dc.relation.referencesPetris, G., Petrone, S., & Campagnoli, P. (2009). Dynamic Linear Models with R. Retrieved from http://www.springer.com/statistics/statistical+theory+and+methods/book/978-0-387-77237- 0spa
dc.relation.referencesR-Devleopment-Core-Team. (2014). A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. Retrieved from http://www.rproject.org/spa
dc.relation.referencesSimchi-Levi, D., Kaminski, P., & Simchi-Levi, E. (2008). Designing and Managing the Supply Chain. (McGraw-Hill, Ed.) (3rd ed.). New York: McGraw-Hill.spa
dc.relation.referencesWang, J., & Hu, J. (2015). A robust combination approach for short-term wind speed forecasting and analysis - Combination of the ARIMA (Autoregressive Integrated Moving Average), ELM (Extreme Learning Machine), SVM (Support Vector Machine) and LSSVM (Least Square SVM) forecasts usi. Energy, 93, 41–56. http://doi.org/10.1016/j.energy.2015.08.045spa
dc.relation.referencesZotteri, G., & Kalchschmidt, M. (2007). A model for selecting the appropriate level of aggregation in forecasting processes. International Journal of Production Economics, 108(1-2), 74–83. http://doi.org/10.1016/j.ijpe.2006.12.030spa
dc.rightsDerechos reservados - Universidad Nacional de Colombiaspa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-NoComercial-SinDerivadas 4.0 Internacionalspa
dc.rights.licenseAtribución-NoComercial-SinDerivadas 4.0 Internacionalspa
dc.rights.spaAcceso abiertospa
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/spa
dc.subject.ddcMatemáticas::Probabilidades y matemáticas aplicadasspa
dc.subject.proposalBúsqueda tabúspa
dc.subject.proposalMetaheuristicseng
dc.subject.proposalForecast Combinationeng
dc.subject.proposalPronosticospa
dc.subject.proposalProgramación evolutivaspa
dc.subject.proposalRegresión bayesianaspa
dc.titleForecast combination using Optimization techniquesspa
dc.title.alternativeForecast Combinationspa
dc.typeLibrospa
dc.type.coarhttp://purl.org/coar/resource_type/c_2f33spa
dc.type.coarversionhttp://purl.org/coar/version/c_970fb48d4fbd8a85spa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/bookspa
dc.type.redcolhttp://purl.org/redcol/resource_type/LIBspa
dc.type.versioninfo:eu-repo/semantics/publishedVersionspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
9789584885074.pdf
Tamaño:
1.18 MB
Formato:
Adobe Portable Document Format
Descripción:
Forecast combination using Optimization techniques

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
3.9 KB
Formato:
Item-specific license agreed upon to submission
Descripción:

Colecciones