Modelo de pronóstico para el producto bruto interno (PBI) en el Perú, periodo 2017-2023
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Nacional de Trujillo
Abstract
Esta investigación tuvo como objetivo determinar el modelo de pronóstico adecuado del Producto Bruto Interno (PBI) en Perú durante el periodo de enero de 2017 a diciembre de 2023. Utilizando la metodología Box-Jenkins, se analizó la tendencia, estacionariedad y estacionalidad de la serie de tiempo, Inicialmente, la serie original no era estacionaria debido a la presencia de una tendencia y estacionalidad, evidenciando un pico en 2022 y una caída en 2020. Al aplicar la primera diferencia regular, la serie se volvió estacionaria. Tras evaluar varios modelos, se identificó el modelo AR(1) SAR(12) MA(1) SMA(12) como el más apropiado para capturar las características cíclicas y volátiles del PBI. Este modelo fue validado estadísticamente, mostrando coeficientes significativos, una distribución normal de los residuos, ausencia de heterocedasticidad e independencia de los errores. Los errores de pronóstico (ECM, EAM, MAPE) fueron bajos, lo que indica una alta precisión y un buen ajuste a los datos reales. Según el modelo, se pronostica un aumento del PBI para 2024, manteniendo la volatilidad histórica, pero con una tendencia alcista
Abstract This research aimed to determine the appropriate forecast model for the Gross Domestic Product (GDP) in Peru during the period from January 2017 to December 2023. Using the Box-Jenkins methodology, the trend, stationarity and seasonality of the series were analyzed. of time, Initially, the original series was not stationary due to the presence of a trend and seasonality, evidencing a peak in 2022 and a drop in 2020. By applying the first regular difference, the series became stationary. After evaluating several models, the AR(1) SAR(12) MA(1) SMA(12) model was identified as the most appropriate to capture the cyclical and volatile characteristics of GDP. This model was statistically validated, showing significant coefficients, a normal distribution of residuals, absence of heteroscedasticity and independence of errors. Forecast errors (ECM, EAM, MAPE) were low, indicating high precision and good fit to actual data. According to the model, an increase in GDP is forecast for 2024, maintaining historical volatility, but with an upward trend
Abstract This research aimed to determine the appropriate forecast model for the Gross Domestic Product (GDP) in Peru during the period from January 2017 to December 2023. Using the Box-Jenkins methodology, the trend, stationarity and seasonality of the series were analyzed. of time, Initially, the original series was not stationary due to the presence of a trend and seasonality, evidencing a peak in 2022 and a drop in 2020. By applying the first regular difference, the series became stationary. After evaluating several models, the AR(1) SAR(12) MA(1) SMA(12) model was identified as the most appropriate to capture the cyclical and volatile characteristics of GDP. This model was statistically validated, showing significant coefficients, a normal distribution of residuals, absence of heteroscedasticity and independence of errors. Forecast errors (ECM, EAM, MAPE) were low, indicating high precision and good fit to actual data. According to the model, an increase in GDP is forecast for 2024, maintaining historical volatility, but with an upward trend
Description
Keywords
Producto bruto interno; pronostico