Pronóstico de la exportación de arándanos en el Perú, periodo 2016-2023
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Nacional de Trujillo
Abstract
El propósito fundamental de este estudio de investigación es identificar el modelo de pronóstico adecuado para describir el comportamiento de las exportaciones mensuales de arándanos en Perú. Se emplearán datos proporcionados por el Banco Central de Reserva del Perú (BCRP) abarcando el período entre enero de 2016 y octubre de 2023 para posteriormente aplicar series de tiempo. Siendo esta una investigación aplicada y longitudinal. La variable de estudio es la exportación mensual de arándanos (millones de US$/mes). Se utilizó la metodología de Box Jenkins y la serie se dividió en: enero 2016 a octubre 2022 para la estimación del modelo y de noviembre 2022 a octubre 2023 para la evaluación del pronóstico, esta metodología estadística se ocupa de identificar y ajustar modelos para series temporales, donde se estimó los 3 modelos, donde se escogió el mejor modelo con menor C.I. y luego se procedió a validarlo los supuestos. Por último, se examinó la predicción, llevando a cabo el procesamiento mediante el programa estadístico Eviews 12 con el fin de comparar los resultados, se llegó a la conclusión de que el modelo de pronóstico más adecuado es un AR(1) MA(1) SMA(12) MA(23) AR(11) SAR(36). La ecuación que lo representa es: ΔŶt = 0.51𝑦𝑡−1 + 0.48𝑦𝑡−11 − 1.04𝜀𝑡−1+ 0.04𝜀𝑡−23 −+ 0.049s𝑦𝑡−36 + 0.25s𝑦𝑡−12 + 𝜀𝑡 después de verificar la idoneidad del modelo, se llevó a cabo la predicción de las exportaciones mensuales desde noviembre de 2023 hasta junio de 2024.
ABSTRACT The primary purpose of this research study is to identify the most suitable forecasting model to describe the evolution of monthly blueberry exports in Peru. Data provided by the Central Reserve Bank of Peru (BCRP) for the period from January 2016 to October 2023 will be employed to later apply time series. This being an applied and longitudinal research. The variable under investigation is the monthly blueberry export (millions of US$/month). The Box Jenkins methodology was utilized, and the data series was divided into two periods: January 2016 to October 2022 for model estimation and November 2022 to October 2023 for forecast evaluation. This statistical methodology deals with identifying and adjusting models for time series, where the 3 models were estimated, where the best model with the lowest C.I. was chosen. and then the assumptions were validated. Finally, the prediction was examined, carrying out the processing using the Eviews 12 statistical program in order to compare the results. It was concluded that the most appropriate forecasting model is a AR(1) MA(1) SMA(12) MA(23) AR(11) SAR(36). The equation representing it is: ΔŶt = 0.51𝑦𝑡−1 + 0.48𝑦𝑡−11 − 1.04𝜀𝑡−1 + 0.04𝜀𝑡−23 −+ 0.049s𝑦𝑡−36 + 0.25s𝑦𝑡−12+ 𝜀𝑡. After verifying the model's suitability, monthly export predictions were made from November 2023 to June 2024.
ABSTRACT The primary purpose of this research study is to identify the most suitable forecasting model to describe the evolution of monthly blueberry exports in Peru. Data provided by the Central Reserve Bank of Peru (BCRP) for the period from January 2016 to October 2023 will be employed to later apply time series. This being an applied and longitudinal research. The variable under investigation is the monthly blueberry export (millions of US$/month). The Box Jenkins methodology was utilized, and the data series was divided into two periods: January 2016 to October 2022 for model estimation and November 2022 to October 2023 for forecast evaluation. This statistical methodology deals with identifying and adjusting models for time series, where the 3 models were estimated, where the best model with the lowest C.I. was chosen. and then the assumptions were validated. Finally, the prediction was examined, carrying out the processing using the Eviews 12 statistical program in order to compare the results. It was concluded that the most appropriate forecasting model is a AR(1) MA(1) SMA(12) MA(23) AR(11) SAR(36). The equation representing it is: ΔŶt = 0.51𝑦𝑡−1 + 0.48𝑦𝑡−11 − 1.04𝜀𝑡−1 + 0.04𝜀𝑡−23 −+ 0.049s𝑦𝑡−36 + 0.25s𝑦𝑡−12+ 𝜀𝑡. After verifying the model's suitability, monthly export predictions were made from November 2023 to June 2024.
Description
Keywords
Exportación mensual de Arándanos