Macroeconomic Forecasting With Mixed-Frequency Data

• DOI: 10.1198/073500108000000015
• Publication Date: 2008-10-02T00:17:31Z
• Authors: Michael P Clements; Ana Beatriz Galvão
• Keywords: Nowcasting; Quarter (Canadian coin)

Many macroeconomic series, such as U.S. real output growth, are sampled quarterly, although potentially useful predictors often observed at a higher frequency. We look whether mixed data-frequency sampling (MIDAS) approach can improve forecasts of growth. The MIDAS specification used in the comparison uses novel way including an autoregressive term. find that use monthly data on current quarter leads to significant improvement forecasting and next is effective exploit compared with alternative methods.