Research
Publications
Education

Riccardo Faini CEIS Seminars

Structured Regularization of High-Dimensional Panel Vector Autoregressions
May, 07th 2021 (16:00-17:00)
CEIS Tor Vergata – TEAMS Webinar

Stephen Smeekes (Maastricht University)

Riccardo Faini CEIS Webinars

Registration form - no later than Thursday

joint with L. Lieb and M. Staudigl

We consider estimation of large panel vector autoregressions using regularization techniques. Rather than relying on standard sparsity assumptions which are not likely to be applicable in panel time series (for example in macroeconomics), our regularization exploits structured sparsity that exists naturally by similarity of the dynamics within and between groups in the panel. Moreover, our estimation method endogenously provides a measure of the strength of the (dynamic) interconnectedness between the groups. We investigate the finite sample properties of our method in an extensive simulation study, and apply our method to study the effects of carbon taxes in European countries. For this purpose we also provide an accompanying identification scheme that allows for structural analysis.

 

Back