Minnesota-Type Adaptive Hierarchical Priors for Large Bayesian VARs

Joshua Chan (2021)
International Journal of Forecasting, 37(3): 1212-1226
[ Journal Version | Working Paper | Code ]

This code estimates and computes forecasts from large Bayesian VARs with 3 different hierarchical shrinkage priors on the VAR coefficients: the proposed Minnesota-type normal-gamma prior, the normal-gamma prior and a data-based Minnesota prior.

The application is illustrated using US datasets of 23 and 30 macroeconomic variables at quarterly frequency.