Title: The Identifying information in the forecast error variance: An application to uncertainty shocks
Authors: Andrea Carriero - QMUL (United Kingdom)
Alessio Volpicella - University of Surrey (United Kingdom) [presenting]
Abstract: A Structural Vector Autoregression identification scheme is developed based on inequality constraints on the Forecast Error Variance decomposition. We characterise the topological properties of this approach and provide algorithms for estimation and inference. We use this strategy to investigate the effects of uncertainty shocks on the economy by allowing for endogeneity, disentangling different sources of uncertainty, and separating uncertainty from pure financial shocks. Monte-Carlo exercises illustrate the effectiveness of this approach. Using US data, we find that some macro variables have a significant contemporaneous feedback effect on financial uncertainty, and overlooking this channel can lead to distortions in the estimated effects of uncertainty on the economy. Also, ignoring that uncertainty has heterogeneous sources biases the estimation. Finally, omitting the endogenous features of financial uncertainty leads to underestimating the effects of financial shocks on the economy. The relationship between these results and recent theoretical contributions is discussed.