Page 1 of 1
Posted: Wed May 31, 2017 6:01 pm
Dear friends, I have encountered a simple problem about BVAR.
I compute BVAR directly in Eviews.
Firstly, if I want to compute the 90-percent error bands in dashed lines, is there any option for it in Eviews?
I use litterman to compute it.
Thanks for any help in advance!
Posted: Wed Jun 07, 2017 4:34 pm
Try the LBVAR add-in.
Posted: Tue Aug 22, 2017 1:58 am
I am estimating a large BVAR in EViews with 18 variables, a 12 lag structure and N-W priors.
Has anyone else encountered EViews crashing with this level of computation or is it a problem on my end?
Posted: Tue Aug 22, 2017 2:27 am
try to use the LBVAR add-in. It estimates LARGE BVAR model.
Posted: Wed Aug 23, 2017 12:24 am
Thanks for recommending the LBVAR add in. Is there any literature around the add in? I have read the paper that the add in is based on and I am a little lost as to the 'number of horizons' and 'number of MC draws' in the specification.
Posted: Wed Aug 30, 2017 3:38 am
I have now found the documentation! Apologies - Abi