How accurate are professional forecasts in Asia? Evidence from ten countries

Abstract

This paper assesses the performances of professional GDP growth and inflation forecasts for ten Asian economies for the period 1995–2012. We evaluate the accuracy of the forecasts, and test for unbiasedness and efficiency. Our results show that (i) forecast errors are large for most of the countries, but there are big differences between countries; (ii) forecasts improve slowly as the forecast horizon shortens, which helps to explain the magnitudes of the forecast errors; (iii) GDP growth forecasts underreact to economic news but inflation forecasts are mostly efficient; (iv) the sizes of forecast biases vary widely between countries, with a tendency for inflation to be overestimated; and (v) forecasts have value in predicting the direction of change.