Using Time Distance to select forecasting models of in ation

This paper applies time distance, suggested by Granger and Jeon (1997), to models of in ation. The commonly used metric of forecast performance, mean squared forecasting error, measures how vertically close the forecast is to the series to be predicted; time distance instead measures the leading and lagging properties, often leading to significantly di erent conclusions about relative performance. The e ect of data snooping is examined with White's (1997) reality check bootstrapping technique. This paper speci cally addresses the problem of using measures of core in ation. The core rate is identi ed using statistical dominance and risk-reduction dominance. Time distance selects commodity price as a leading indicator, in contrast with the results of using mean squared forecast error.