No. 271 Un-truncating VARs

By Ferre De Graeve and Andreas Westermark

 

June 2013

 

Abstract

Macroeconomic research often relies on structural vector autoregressions to uncover empirical regularities. Critics argue the method goes awry due to lag truncation: short lag-lengths imply a poor approximation to DSGE-models. Empirically, short lag-length is deemed necessary as increased parametrization induces excessive uncertainty. The paper shows that this argument is incomplete. Longer lag-length simultaneously reduces misspeci.cation, which in turn reduces variance. For data generated by frontier DSGE-models long-lag VARs are feasible, reduce bias and variance, and have better coverage. Thus, contrary to conventional wisdom, the trivial solution to the critique actually works.

Keywords:

VAR, SVAR, Lag-length, Truncation

JEL:

C18, E37

Last reviewed

Content expert

Contact content expert

Fill in the information

To minimize automated spam, please answer the question in the box below.

7 + 7 ?