Hi,
is it possible to choose a prior mean different than zero in the Ko-Ko Minnesota and Normal-Wishart priors? I am estimating a VAR in levels and hence would like to adopt the original Litterman prior where the prior mean on each variable's first lag is 1 instead of 0.
Thanks.
Choosing a prior mean in BVAR
Moderators: EViews Gareth, EViews Moderator, EViews Esther
-
- EViews Developer
- Posts: 149
- Joined: Fri Sep 03, 2010 7:57 am
Re: Choosing a prior mean in BVAR
Please take a look at the latest BVAR addin (note: changes are updated on viewtopic.php?f=23&t=3295). You can now change the prior mean for the coefficient parameter BETA.
Re: Choosing a prior mean in BVAR
Hi Esther, thanks for your reply.
They way the code is written now it sets the prior mean of all coefficients to beta0. Actually, what I meant is if it would be possible to just change the prior mean on the first own lag of each variable to 1(or to whichever other value, 0.9, 0.8, etc.).
They way the code is written now it sets the prior mean of all coefficients to beta0. Actually, what I meant is if it would be possible to just change the prior mean on the first own lag of each variable to 1(or to whichever other value, 0.9, 0.8, etc.).
Who is online
Users browsing this forum: No registered users and 17 guests