Structural Vector Autoregressive Models with More Shocks Than Variables Identified via Heteroskedasticity

Aufsätze referiert extern - Web of Science

Helmut Lütkepohl

In: Economics Letters 195 (2020), 109458, 4 S.


In conventional structural vector autoregressive models it is assumed that there are at most as manystructural shocks as there are variables in the model. It is pointed out that heteroskedasticity can beused to identify more shocks than variables. Results are provided that allow a researcher to assesshow many shocks can be identified from specific forms of heteroskedasticity.

JEL-Classification: C32
Keywords: Structural vector autoregression, Identification through heteroskedasticity, Structural shock