Quantile-based inflation risk models

This paper proposes a new approach to extract quantile-based inflation risk measures using Quantile Autoregressive Distributed Lag Mixed-Frequency Data Sampling (QADL-MIDAS) regression models. We compare our models to a standard Quantile Auto-Regression (QAR) model and show that it delivers better quantile forecasts at several forecasting horizons. We use the QADL-MIDAS model to construct inflation risk measures proxying for uncertainty, third-moment dynamics and the risk of extreme inflation realizations. We find that these risk measures are linked to the future evolution of inflation and changes in the effective federal funds rate.