BVAR Koop-Korobilis Eviews 10
Posted: Mon Oct 21, 2019 3:56 am
Hello,
I am a Ph.D. student and I am new at the Eviews 10. I would like to run the Koop-Korobilis (2009) (Litterman/Minnesota) prior. I have seen previous discussions in the forum, however the new version seems quite different from the old ones (Eviews 7). I have seen a study that sets Mu1=0, lambda1=0.5, lambda2=0.5, lambda3=100.Do these correspond to the equivalent hyperparameters of Koop-Korobilis? Also, do we impose the VAR as diagonal or univariate AR? Finally, (according to Koop-Korobilis) the first own lag of each variable is set at 0.9 (α min). Should we impose this restriction in some cerτain way or is it set as default by the prior?
Thanks in advance,
I am a Ph.D. student and I am new at the Eviews 10. I would like to run the Koop-Korobilis (2009) (Litterman/Minnesota) prior. I have seen previous discussions in the forum, however the new version seems quite different from the old ones (Eviews 7). I have seen a study that sets Mu1=0, lambda1=0.5, lambda2=0.5, lambda3=100.Do these correspond to the equivalent hyperparameters of Koop-Korobilis? Also, do we impose the VAR as diagonal or univariate AR? Finally, (according to Koop-Korobilis) the first own lag of each variable is set at 0.9 (α min). Should we impose this restriction in some cerτain way or is it set as default by the prior?
Thanks in advance,