Addressing COVID-19 Outliers in BVARs with Stochastic Volatility

Abstract The COVID-19 pandemic has led to enormous data movements that strongly affect parameters and forecasts from standard Bayesian vector autoregressions (BVARs). To address these issues, we propose BVAR models with outlier-augmented stochastic volatility (SV) combine transitory persistent changes in volatility. resulting density are much less sensitive outliers the than BVARs. Predictive Bayes factors indicate our SV model provides best fit for period, as well earlier subsamples of high In historical forecasting, schemes fare at least a conventional model.