Prediction of the export price of pitahaya in Ecuador using the SARIMAX model

This study aims to predict the export price of pitahaya in Ecuador through the application of the SARIMAX model, using variables associated with the historical behavior of prices and monthly export volume, to generate useful information for producers’ decision-making. The research adopted a quantita...

Descripción completa

Detalles Bibliográficos
Autores: García Bailón, Elian Jesús, Párraga Muñoz, Sonia Monserrate
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2026
País:Ecuador
Institución:Universidad San Gregorio de Portoviejo
Repositorio:Revista Universidad San Gregorio de Portoviejo
Idioma:español
OAI Identifier:oai:ojs.pkp.sfu.ca:article/3862
Acceso en línea:https://revista.sangregorio.edu.ec/index.php/REVISTASANGREGORIO/article/view/3862
Access Level:acceso abierto
Palabra clave:Dragon fruit
Price forecasting
Time series
Agricultural exports
Ecuador
Pitahaya
Predicción de precios
Series temporales
Exportación agrícola
Descripción
Sumario:This study aims to predict the export price of pitahaya in Ecuador through the application of the SARIMAX model, using variables associated with the historical behavior of prices and monthly export volume, to generate useful information for producers’ decision-making. The research adopted a quantitative approach, with a predictive scope and a non-experimental design. Monthly time series of the average price per kilogram of pitahaya were analyzed for the period from September 2021 to May 2025, based on records from producers in the province of Manabí. A SARIMAX model with an annual seasonal component was estimated, incorporating monthly export volume as an exogenous variable. Model performance was evaluated using MAE, RMSE, MAPE, and information criteria. The model made it possible to identify seasonal patterns and price variations associated with exportable supply. The inclusion of the exogenous variable provided greater sensitivity to market changes, especially during periods of higher product availability. The metrics obtained showed acceptable performance for predictive purposes, with a MAPE of 12.45%. It is concluded that the SARIMAX model constitutes a useful statistical tool for anticipating price fluctuation scenarios in pitahaya exports. Its application may contribute to improving production and commercial planning, although the incorporation of additional external variables is recommended to strengthen its predictive accuracy.