Large Hybrid Time-Varying Parameter VARs

Time-varying parameter VARs with stochastic volatility are routinely used for structural analysis and forecasting in settings involving a few endogenous variables. Applying these models to high-dimensional datasets has proved be challenging due intensive computations over-parameterization concerns. We develop an efficient Bayesian sparsification method class of we call hybrid TVP-VARs—VARs time-varying parameters some equations but constant coefficients others. Specifically, each equation, the new automatically decides whether VAR contemporaneous relations among variables or time-varying. Using U.S. various dimensions, find evidence that some, not all, time varying. The large TVP-VAR also forecasts better than many standard benchmarks. Supplementary materials this article available online.