Predictive Capabilities of Macroeconomic Exchange Rate Determinants

In this paper, the Dynamic Bayesian Model Averaging (DMA) algorithm is used to establish the critical determinants of the nominal exchange rates of 5 currencies: CAD, EUR, GBP, CHF and JPY against the US dollar. My results indicate that the set of essential exchange rate predictors can substantially differ in time. The exchange rate's lagged value always remains a vital variable indicating significant persistence in the exchange rate time series. Furthermore, the PPP rate, Terms of Trade (TOT) and output per worker are also variables with high Posterior Inclusion Probabilities among the analysed countries. My results show that macroeconomic fundamentals are not leading indicators of the exchange rates. As a result, to outperform the random walk (naive) forecast of the exchange rate using the macroeconomic fundamentals, high quality of the explanatory variables' forecast is required