🤖 AI Summary
This study addresses the computational inefficiencies and high autocorrelation commonly encountered in traditional Bayesian inference for high-dimensional structural vector autoregressive (SVAR) models with symbolic constraints. The authors propose a reparameterization-based Hamiltonian Monte Carlo (HMC) algorithm that unifies inequality identification constraints—such as shape, ordering, and elasticity bounds—with zero restrictions into a continuously differentiable mapping. This approach enables, for the first time, differentiable handling of complex mixed constraints within SVAR estimation. Empirical results demonstrate that the method substantially improves posterior sampling efficiency, yielding lower autocorrelation, higher effective sample sizes, and reduced computation time. Consequently, the proposed framework enhances both the stability and scalability of Bayesian inference in high-dimensional SVAR models.
📝 Abstract
We propose a new approach to inference in tightly identified and large-scale structural vector autoregressions based on a reparameterization that enables imposing identifying inequality restrictions through continuously differentiable mappings. Permitted inequality restrictions include shape and ranking restrictions as well as bounds on economically relevant elasticities, and the approach is also able to accommodate zero restrictions in a straightforward manner. We implement a Hamiltonian Monte Carlo algorithm and show how the posterior density can be rapidly evaluated under the reparameterization, thus facilitating inference in high-dimensional settings. Two empirical applications demonstrate that our approach tends to result in lower serial dependence in Markov chains, larger effective sample sizes and reduced computation time relative to existing methods.