Forecasting Inflation with the New Keynesian Phillips Curve: Frequency Matters

Abstract

We show that the New Keynesian Phillips Curve (NKPC) outperforms standard benchmarks in forecasting U.S. ination once frequency-domain information is taken into account.We do so by decomposing the time series (of ination and its predictors) into several frequency bands and forecasting separately each frequency component of ination. The largest statistically signicant forecasting gains are achieved with a model that forecasts the lowest frequency component of ination (corresponding to cycles longer than 16 years) exibly using information from all frequency components of the NKPC ination predictors. Its performance is particularly good in the returning to recovery from the Great Recession.