Empirical evidence of the monetary approach to the exchange rate determinants under a fully flexible regime: the case of Mexico

Abstract

The purpose of this paper is to examine the empirical evidence on the evolution of the nominal peso-dollar exchange rate based on the monetary approach under a fully flexible exchange rate regime. We use a standardised framework where the uncovered interest rate (UIP) and the purchase power parities (PPP), flexible prices, and a typical demand for real money balances determine prices in the long run. Once we identify that time series of the nominal exchange rate and the fundamental macroeconomic variables are non-stationary, we estimate a vector error-correction model (VECM) and, for comparative purposes, an ARIMA-EGARCH model and an ARIMA-EGARCH model with monetary approach. Models' assessment based on the post estimation results shows that the model with the lowest HRMSE for all the steps-ahead forecast is the ARIMA-EGARCH model followed by the VECM model. Likewise, the lowest HMAE for the first three-steps ahead forecast is the VECM model, followed by the ARIMA-EGARCH model.