Density forecasting using Bayesian global vector autoregressions with stochastic volatility

B-Tier
Journal: International Journal of Forecasting
Year: 2016
Volume: 32
Issue: 3
Pages: 818-837

Score contribution per author:

2.011 = (α=2.01 / 1 authors) × 1.0x B-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

This paper develops a Bayesian global vector autoregressive model with stochastic volatility. Three variants of the stochastic volatility are implemented in an attempt to improve the existing homoscedastic framework. Our baseline model assumes that the variance–covariance matrix is driven by a set of idiosyncratic, country-specific and regional factors. In contrast, the second specification adopted implies that the error variance of each equation is determined by an independent stochastic process. The final specification assumes that the country-specific volatility follows a single factor, which leads to significant computational gains. Considering a range of competing models, we forecast a large panel of macroeconomic variables and find that the stochastic volatility influences the predictive accuracy in three ways. First, it helps to improve the overall predictive fit of our model. Second, it helps to make the model more resilient to outliers and economic crises. Finally, taking a regional stance reveals that the forecasts in developing economies tend to profit more from the use of the stochastic volatility.

Technical Details

RePEc Handle
repec:eee:intfor:v:32:y:2016:i:3:p:818-837
Journal Field
Econometrics
Author Count
1
Added to Database
2026-02-02