A Bayesian VAR Benchmark for COMPASS

Abstract

We estimate a Bayesian VAR analogue to the Bank of England’s DSGE model (COMPASS) and assess their relative performance in forecasting GDP growth and CPI inflation in real time between 2000 and 2012. We find that the BVAR outperformed COMPASS when forecasting both GDP and its expenditure components. In contrast, the performance of these models was similar when forecasting CPI. We also find that, despite underpredicting inflation at most forecast horizons, the BVAR density forecasts outperformed those of COMPASS. Both models overpredicted GDP growth at all forecast horizons, but the BVAR outperformed COMPASS at forecast horizons up to one year ahead. The BVAR’s point and density forecast performance is also comparable to that of a Bank of England in-house statistical suite for both GDP and CPI inflation and to the Inflation Report projections. Our results are broadly consistent with the findings of similar studies for other advanced economies.

Keywords: Forecasting, Bayesian VARs, macro-modelling

JEL Classification: C53, E12, E17