460 – Forecasting and Official Statistics
Testing for and Estimating Arbitrarily Time-varying Forecast Bias
Neil Ericsson
Board of Governors of the Federal Reserve System
Impulse indicator saturation (IIS) generalizes the standard Mincer-Zarnowitz test of time-invariant forecast bias by allowing for arbitrarily time-varying forecast bias. Using both approaches, the current paper analyzes potential biases in different U.S. government agencies' one-year-ahead forecasts of U.S. gross federal debt over 1984-2012. Standard tests typically fail to detect biases, whereas IIS detects economically large and highly significant time-varying biases, particularly at turning points in the business cycle. IIS defines a generic procedure for examining forecast properties; it explains why standard tests fail to detect bias; and it provides a mechanism for potentially improving forecasts.