Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 290 - Advanced Bayesian Topics (Part 3)
Type: Contributed
Date/Time: Wednesday, August 11, 2021 : 1:30 PM to 3:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #318660
Title: Sparse Bayesian High-Dimensional Vector Autoregressions
Author(s): Rui Meng*
Companies: Lawrence Berkeley National Laboratory
Keywords: Bayesian vector autoregression; Shrinkage prior; high dimension
Abstract:

Vector autoregressive (VAR) models have been widely used for modeling the temporal dependence of multivariate time series. Global-local priors are increasingly used to induce shrinkage in such models. In this article, we introduce both element-wise and lag-wise shrinkage on the VAR with stochastic volatility case via three-parameter beta prior. And we propose a new Bayesian estimation procedure for possibly high-dimensional VARs. We show that our proposed framework yields more precise estimates of model parameters and performs well in applications to out-of-sample forecasting.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program