Title: Asymmetries and unemployment rate forecasts
Authors: Simon van Norden - HEC Montreal (Canada) [presenting]
John Galbraith - McGill University (Canada)
Abstract: Asymmetries in unemployment dynamics have been observed in the historical time series of unemployment rates in a number of countries, including the United States. Patterns in the corresponding unemployment rate forecasts are studied. We consider conditions under which optimal forecasts will display asymmetry, and the patterns that we would expect to see in forecasts or forecast errors at increasingly long horizons. Using data from the U.S. Survey of Professional Forecasters, we examine the patterns associated with informational rigidity and test the hypothesis, suggested by a simple model of asymmetric response to an underlying causal factor, that asymmetry in forecast errors will be non-decreasing with forecast horizon.