A panel threshold VAR with stochastic volatility-in-mean model: an application to the effects of financial and uncertainty shocks in emerging economies

Abstract

ABSTRACT This article proposes a panel threshold VAR with stochastic volatility-in-mean model. The volatility process is given by the average volatility of the structural shocks affecting some selected variables and is interpreted as a measure of macroeconomic uncertainty. The posterior is approximated using an adaptive Metropolis-Within-Gibbs algorithm. We apply the method by analysing the effects of uncertainty and financial shocks in good times and financial distress in emerging economies. In our sample, we find that, under normal financial conditions, financial shocks play a slightly more pervasive role than uncertainty shocks. However, under financial stress periods, uncertainty shocks become more important than financial shocks.