The unbeatable random walk in exchange rate forecasting: Reality or myth?

It is demonstrated that the conventional monetary model of exchange rates can (irrespective of the specification, estimation method or the forecasting horizon) outperform the random walk in out-of-sample forecasting if forecasting power is measured by direction accuracy and profitability. Claims of outperforming the random walk in terms of the root mean square error are false because they are typically based on the introduction of dynamics, hence a random walk component, commonly without testing for the statistical significance of the difference between root mean square errors. And even if proper hypothesis testing reveals that a dynamic model outperforms the random walk, this amounts to beating the random walk by a random walk with the help of some explanatory variables. The failure of conventional macroeconomic models to outperform the random walk in terms of the root mean square error should be expected rather than considered to be a puzzle