Estimation and Inference of the Forecast Error Variance Decomposition for Set-Identified SVARs

Seminar summary

We study the Structural Vector Autoregressions (SVARs) that impose internal and external restrictions to set-identify the Forecast Error Variance Decomposition (FEVD). We make the following contributions. First, we characterize the endpoints of the FEVD as the extreme eigenvalues of a symmetric reduced-form matrix. A consistent plug-in estimator naturally follows. Second, we use the perturbation theory to prove that the endpoints of the FEVD are differentiable with respect to the reduced-form parameters. Third, we rely on inference for eigenvalues to construct confidence intervals that are uniformly consistent in level and have asymptotic robust Bayesian credibility. A Monte-Carlo exercise demonstrates the approach properties in finite samples. A credit supply application illustrates our toolkit.

How to attend this seminar

This seminar will take place on Wednesday 21 February 2024 at 3pm.

We welcome you to join us at our Colchester campus in room EBS.2.1.

If you are unable to make it in person you can also join us online.

Speaker bio

Dr Alessio Volpicella 

Alessio Volpicella is a Senior Lecturer of Economics at the University of Surrey, where he is also affiliated with the Centre for International Macroeconomic Studies and a Fellow of the Artificial Intelligence Institute. He holds an Economics PhD from the Queen Mary University of London (2020); his research includes time series analysis, macroeconometrics and Bayesian Econometrics. He publishes in top field journals such as Journal of Business and Economic Statistics and Quantitative Economics.
He has collaborated with the European Central Bank, the Bank of England, BNP Paribas, and the Department of Business and Trade.