COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES
In the food industry, inventory models that provide adequate prediction of demand are useful for optimum supply management, especially in perishable products such as dairy products, because of their short shelf life and the importance of their quality related to health. However, some small and mediu...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | Spanish |
| Published: |
Universidad Nacional de Colombia, sede Medellín
2017-07-01
|
| Series: | Revista de la Facultad de Ciencias |
| Subjects: | |
| Online Access: | https://revistas.unal.edu.co/index.php/rfc/article/view/66085 |
| _version_ | 1818408532239712256 |
|---|---|
| author | Marisol Valencia Cárdenas Victor Alfonso Osorno Vásquez Juan Carlos Salazar Uribe |
| author_facet | Marisol Valencia Cárdenas Victor Alfonso Osorno Vásquez Juan Carlos Salazar Uribe |
| author_sort | Marisol Valencia Cárdenas |
| collection | DOAJ |
| description | In the food industry, inventory models that provide adequate prediction of demand are useful for optimum supply management, especially in perishable products such as dairy products, because of their short shelf life and the importance of their quality related to health. However, some small and medium Colombian companies do not have the technology and capabilities to do forecasts of their products, which is very important for production and inventory planning. In this paper we propose a comparison of the precision of forecasts of individual statistical models and combinations between them, using a multi-product algorithm for the combinations applied to a case study of sales of the dairy sector. It is found that an individual model of Bayesian regression with innovation is a very good alternative in the case studied, as well as two of the combination techniques used. |
| first_indexed | 2024-12-14T09:45:13Z |
| format | Article |
| id | doaj.art-a9fb6e25aa624158a70337a57079939b |
| institution | Directory Open Access Journal |
| issn | 0121-747X 2357-5549 |
| language | Spanish |
| last_indexed | 2024-12-14T09:45:13Z |
| publishDate | 2017-07-01 |
| publisher | Universidad Nacional de Colombia, sede Medellín |
| record_format | Article |
| series | Revista de la Facultad de Ciencias |
| spelling | doaj.art-a9fb6e25aa624158a70337a57079939b2022-12-21T23:07:40ZspaUniversidad Nacional de Colombia, sede MedellínRevista de la Facultad de Ciencias0121-747X2357-55492017-07-016212414010.15446/rev.fac.cienc.v6n2.6608546913COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUESMarisol Valencia Cárdenas0Victor Alfonso Osorno Vásquez1Juan Carlos Salazar Uribe2Institución Universitaria Tecnológico de AntioquiaUniversidad Nacional de ColombiaUniversidad Nacional de Colombia, Sede MedellínIn the food industry, inventory models that provide adequate prediction of demand are useful for optimum supply management, especially in perishable products such as dairy products, because of their short shelf life and the importance of their quality related to health. However, some small and medium Colombian companies do not have the technology and capabilities to do forecasts of their products, which is very important for production and inventory planning. In this paper we propose a comparison of the precision of forecasts of individual statistical models and combinations between them, using a multi-product algorithm for the combinations applied to a case study of sales of the dairy sector. It is found that an individual model of Bayesian regression with innovation is a very good alternative in the case studied, as well as two of the combination techniques used.https://revistas.unal.edu.co/index.php/rfc/article/view/66085Estadística y probabilidadevaluación de modelos y selecciónmétodos de pronósticoestadística Bayesiana. |
| spellingShingle | Marisol Valencia Cárdenas Victor Alfonso Osorno Vásquez Juan Carlos Salazar Uribe COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES Revista de la Facultad de Ciencias Estadística y probabilidad evaluación de modelos y selección métodos de pronóstico estadística Bayesiana. |
| title | COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES |
| title_full | COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES |
| title_fullStr | COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES |
| title_full_unstemmed | COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES |
| title_short | COMPARATIVE OF FORECASTING MODELS: CLASSICAL, BAYESIAN, AND COMBINATION TECHNIQUES |
| title_sort | comparative of forecasting models classical bayesian and combination techniques |
| topic | Estadística y probabilidad evaluación de modelos y selección métodos de pronóstico estadística Bayesiana. |
| url | https://revistas.unal.edu.co/index.php/rfc/article/view/66085 |
| work_keys_str_mv | AT marisolvalenciacardenas comparativeofforecastingmodelsclassicalbayesianandcombinationtechniques AT victoralfonsoosornovasquez comparativeofforecastingmodelsclassicalbayesianandcombinationtechniques AT juancarlossalazaruribe comparativeofforecastingmodelsclassicalbayesianandcombinationtechniques |