After MH iterations, the computations can get stuck on marginal_density.m because of a log(0).
If you have this problem, you can use this fix: using determinant properties, replace log(detSIGMA) with logdetSIGMA that is defined as
logdetSIGMA = 2*log(det(chol(SIGMA)));
See attached file, it's a modification of the 4.2 version of Dynare.
Hope this help!
Gilles
Dynare forums
Forums for asking questions, posting comments and uploading examples related to Dynare
If calculations after MH take forever
Forum rules
This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location (https://forum.dynare.org) where you will have to reset your password.
This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location (https://forum.dynare.org) where you will have to reset your password.
4 posts
• Page 1 of 1
If calculations after MH take forever
by gbel » Wed Apr 06, 2011 4:07 pm
Last edited by gbel on Wed Apr 06, 2011 7:22 pm, edited 1 time in total.
- gbel
- Posts: 33
- Joined: Thu Sep 04, 2008 8:34 pm
- Location: Québec
Re: If calculations after MH take forever
by gbel » Wed Apr 06, 2011 7:21 pm
A better idea is:
logdetSIGMA = 2*sum(log(diag(chol(SIGMA))));
That way, the estimated model can grow even more before the loss of precision becomes a problem.
I updated the marginal_density.m file.
Gilles
logdetSIGMA = 2*sum(log(diag(chol(SIGMA))));
That way, the estimated model can grow even more before the loss of precision becomes a problem.
I updated the marginal_density.m file.
Gilles
- Attachments
-
marginal_density.m- (4.63 KiB) Downloaded 486 times
- gbel
- Posts: 33
- Joined: Thu Sep 04, 2008 8:34 pm
- Location: Québec
Re: If calculations after MH take forever
by olayenidynare » Thu Jul 21, 2011 11:27 am
This is so useful. The estimation that used to take about 60 mins takes less than 12 mins!!!!!
Many thanks for sharing this file. But what if there is no log(0) problem? Is the same modified file still useful?
Many thanks for sharing this file. But what if there is no log(0) problem? Is the same modified file still useful?
- olayenidynare
- Posts: 22
- Joined: Sat Feb 21, 2009 3:17 pm
Re: If calculations after MH take forever
by gbel » Fri Jul 22, 2011 8:06 pm
Yes.
The formula is exact (you can check for fun). In fact, it's more precise for all computations.
Gilles
The formula is exact (you can check for fun). In fact, it's more precise for all computations.
Gilles
- gbel
- Posts: 33
- Joined: Thu Sep 04, 2008 8:34 pm
- Location: Québec
4 posts
• Page 1 of 1
Return to Dynare contributions and examples
Who is online
Users browsing this forum: No registered users and 2 guests