Modelamiento del precio de la papa criolla en el departamento de Cundinamarca por medio de series de tiempo y modelos dinámicos
Fecha
2021-09-20Registro en:
Díaz, M. E., Cruz, E. A., Pineda, W. D. (2021). Modelamiento del precio de la papa criolla en el departamento de Cundinamarca por medio de series de tiempo y modelos dinámicos. Comunicaciones en Estadística, 14(1), 31-52.
reponame:Repositorio Institucional Universidad Santo Tomás
instname:Universidad Santo Tomás
Autor
Díaz Sosa, María Eliana
Institución
Resumen
The objective of the following research is to evaluate the behavior and forecast of the price of the papa criolla in the department of Cundinamarca, according to the climate factors from January 2012 to April 2018. Therefore, analysis based on time series (ARIMA, ARIMAX) and, on the other hand, dynamic linear models (with and without variables) were taken into consideration. In the models worked, the climate conditions of the area in question were used as variables, to which a data imputation method was applied due to the absence of information. After that, they were grouped into three factors constructed by Time Series Factor Analysis (TSFA). Finally, the indicators of the four models were compared, concluding that the ARIMA and ARIMAX models generate the best predictions regarding the price of papa criolla in the department of Cundinamarca.