Abstract
Using several samples of forward exchange rate forecasts for the British pound vs. the US dollar, this article explores the post-sample predictive performance of adjusting the forecasts for recent empirical bias. Numerical accuracy is assessed via both parametric and nonparametric tests, and directional properties are also evaluated. The evidence suggests that simple linear adjustments can yield significant improvements in predictive accuracy, even if the measured bias in the original forecasts is not statistically significant.
Notes
1 Founded in 1989, Consensus Economics provides forecasting data for numerous macroeconomic variables including inflation, interest rates and foreign exchange rates for more than 70 countries. Consensus Economics polls more than 250 economists from around the world in order to obtain their foreign exchange consensus forecast. Contributors to the foreign exchange forecasts include economists from the Bank of America, JP Morgan, Goldman Sachs and a plethora of other reputable institutions. Source: http://www.consensuseconomics.com/what_are_consensus_forecasts.htm