Copyright (c) 2025 Vision Antataura
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This study analyzes the historical series of Panama’s annual inflation rate for the period 1960–2024, with the objective of characterizing its long-term dynamics and generating a 30-year forecast within the context of a fully dollarized economy without a central bank. Using a quantitative, non-experimental, longitudinal design, the analysis applies the Box–Jenkins methodology, complemented by Augmented Dickey–Fuller tests, structural-break identification (Supremum Wald test), cointegration verification, and full residual diagnostics. The inflation series was found to be stationary at level (d = 0), and a significant structural break was identified in 1981. After evaluating multiple specifications, an ARIMA model with a single autoregressive component and a structural break interaction was selected. Diagnostic tests revealed non-normal and heteroskedastic residuals without ARCH effects; therefore, robust (HCE-White) standard errors and a Student’s t-distribution were employed for inference. The resulting model demonstrates strong inflation inertia and heavy-tailed behavior. The 30-year forecast projects a rate of 2.597 % for 2054, with a wide prediction interval that highlights the sensitivity of Panama’s inflation to external shocks. Overall, the model adequately captures historical behavior, but future research should incorporate exogenous global variables to improve predictive accuracy under uncertainty.
Este obra está bajo una licencia de Creative Commons Reconocimiento-NoComercial-CompartirIgual 4.0 Internacional.
