Mixed-frequency Growth-at-Risk with the MIDAS-QR method: Evidence from China

Abstract

High-frequency financial indicators provide more useful information and are efficient at forecasting low-frequency GDP. To this end, we extend the traditional Growth-at-Risk (GaR) framework for mixed frequency data. In this extension, monthly financial indicators are used to forecast quarterly GDP with the mixed data sampling-quantile regression (MIDAS-QR) method. Its ability for high-frequency monitoring of GaR is investigated using Chinese evidence. The evidence shows that our mixed-frequency GaR is promising in terms of good forecasting and nowcasting results, and can offer early warning of GDP downturns.