Forecasting foreign exchange rates in Colombia assuming PPP conditions: Empirical evidence using VAR

Abstract

This document examines exchange rate forecasts during the 1995-2005 period, using a Purchasing Power of Parity Exchange Rate Model (PPPER). Our first finding is that the computed forecasts seem to validate the use of this model under certain conditions given that it performs well in predicting the behavior of the nominal exchange rate. Our second finding included a comparative analysis of out-of-sample forecasts (saving historical data) between the PPP-based forecast models and the Vector Autoregressive (VAR) model. The VAR has a better forecasting performance based on the RMSE, MAE, and U-Theil indicators. MAPE results measured on the first and second month-ahead forecasts indicate that the VAR model performs more poorly than the PPP-based models. © 2018 Tehran University of Medical Sciences. All rights reserved.