Modelo de vectores autorregresivo para la predicción de las exportaciones en el Perú, periodo 1992 – 2024

• Autores: Neira Murrieta, Jhaneth Miyorith, Mejia Guerrero, Jair
• Formato: tesis de grado
• Fecha de Publicación: 2025
• Institución: Universidad Nacional De La Amazonía Peruana
• Repositorio: UNAPIquitos-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.unapiquitos.edu.pe:20.500.12737/11568
• Enlace del recurso: https://hdl.handle.net/20.500.12737/11568
• Nivel de acceso: acceso abierto
• Materia: Modelos econométricos; Predicciones económicas; Estadística; Exportaciones; https://purl.org/pe-repo/ocde/ford#5.02.01; https://purl.org/pe-repo/ocde/ford#5.02.02

The purpose of the study using 386 monthly data from the period 1992:11 to 2024:12 was to compare the forecasting capacity of two multivariate models: the Vector Autoregressive Model (VAR) and the Vector Error Correction Model (VECM). According to the indicators Root Mean Squared Error (RSME) and the average of the errors measured as the percentage of the difference (in absolute value) between the observed and predicted value (Mean Absolute Percentage Error – MAPE), the VECM model reported better performance, since the RMSE was 174 million dollars lower than that of the VAR, which showed 400 million. The MAPE was 50.44% compared to the VAR MAPE of 263%. The Granger causality test shows that there is causality between the study variables except for exports and the multilateral real exchange rate, which do not explain the inflation of trading partners, which makes economic sense.