Working Paper: NBER ID: w19712
Authors: Frank Schorfheide; Dongho Song
Abstract: This paper develops a vector autoregression (VAR) for time series which are observed at mixed frequencies - quarterly and monthly. The model is cast in state-space form and estimated with Bayesian methods under a Minnesota-style prior. We show how to evaluate the marginal data density to implement a data-driven hyperparameter selection. Using a real-time data set, we evaluate forecasts from the mixed-frequency VAR and compare them to standard quarterly-frequency VAR and to forecasts from MIDAS regressions. We document the extent to which information that becomes available within the quarter improves the forecasts in real time.
Keywords: mixed-frequency VAR; forecasting; Bayesian methods; marginal data density
JEL Codes: C11; C32; C53
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| use of within-quarter monthly data (C82) | improvement in forecasting accuracy (C53) |
| switching from QFVAR to MFVAR (C32) | improvement in one-step-ahead nowcast accuracy (C53) |
| MFVAR's utilization of within-quarter monthly information (C32) | substantial improvements in forecast accuracy (C53) |
| MFVAR (C39) | improvement in forecast accuracy during the first month of the quarter (C53) |
| MFVAR (C39) | improvement in forecast accuracy for quarterly frequency variables (C53) |
| MFVAR (C39) | closer tracking of economic downturns during the 2008-2009 Great Recession (F44) |
| percentage differential in forecast accuracy between MFVAR and MIDAS regressions (C32) | consistent across different forecast horizons (C53) |