Can we beat the Random Walk? The case of survey-based exchange rate forecasts in Chile

Abstract

We examine the accuracy of survey-based expectations of the Chilean exchange rate relative to the US dollar. Our out-of-sample analysis reveals that survey-based forecasts outperform the driftless random walk in terms of mean squared prediction error at several forecasting horizons. A similar result is found when precision is measured in a direction-of-change dimension: survey-based forecasts outperform a “pure luck” benchmark at several forecasting horizons. Our findings suggest that survey-based forecasts of the Chilean exchange rate should be considered as a tough benchmark to beat for economic models, tougher indeed than the traditional driftless random walk.