Identification in Structural VAR models with different volatility regimes
Emanuele Bacchiocchi, University of Milan
Download the Paper (PDF format) - December 14, 2011
Tell a colleague about it.
Printing Tips: Select 'print as image' in the Acrobat print dialog if you have trouble printing.
In this paper we study the identification conditions in structural VAR models with different regimes of volatility. We propose a new specification that allows to address identification in the conventional likelihood-based setup. A formal general framework for identification is developped and it is proved that exact-identification assumptions in the standard SVAR literature appear here to be over-identified, and thus subject to statistical inference. The empirical relevance of the methodology is discussed through an empirical application concerning the relationships between term structure of interest rates and output growth.
C01, C13, C30, C51
"Identification in Structural VAR models with different volatility regimes"
UNIMI - Research Papers in Economics, Business, and Statistics.
Statistics and Mathematics.
Working Paper 55.