Hi,
I have a problem with the check plot. What does it mean when I can see only the loglik kernel, but not the log-post?
Thank you so much,
Chiara
Dynare forums
Forums for asking questions, posting comments and uploading examples related to Dynare
check plot
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.
5 posts
• Page 1 of 1
Re: check plot
by StephaneAdjemian » Thu Nov 04, 2010 8:49 am
Dear Chiara,
This happens because in some cases we cannot distinguish the posterior kernel and the likelihood. You will get this kind of plots if the likelihood is picked in a region where the prior is flat.
Best, Stéphane.
This happens because in some cases we cannot distinguish the posterior kernel and the likelihood. You will get this kind of plots if the likelihood is picked in a region where the prior is flat.
Best, Stéphane.
- StephaneAdjemian
- Posts: 429
- Joined: Wed Jan 05, 2005 4:24 pm
- Location: Paris, France.
Re: check plot
by chiara » Thu Nov 04, 2010 12:50 pm
Thank you for your answer.
I am sorry, but I have another stupid question. Is it possible that this problem induces the result of 0 standard deviation in the posterior maximization? Which could be the causes of this problem? What I have to do?
Thank you so much,
Chiara
I am sorry, but I have another stupid question. Is it possible that this problem induces the result of 0 standard deviation in the posterior maximization? Which could be the causes of this problem? What I have to do?
Thank you so much,
Chiara
- chiara
- Posts: 3
- Joined: Tue Nov 02, 2010 10:51 am
Re: check plot
by StephaneAdjemian » Tue Nov 09, 2010 8:18 am
No, these issues are not related... Zero posterior standard deviations are generally a consequence of a failure of the optimization step. You may try the following:
(1) Change the initial condition of the optimization routine (default is the prior mode)
(2) Change the priors
(3) Change the optimization routine
Best, Stéphane.
(1) Change the initial condition of the optimization routine (default is the prior mode)
(2) Change the priors
(3) Change the optimization routine
Best, Stéphane.
- StephaneAdjemian
- Posts: 429
- Joined: Wed Jan 05, 2005 4:24 pm
- Location: Paris, France.
5 posts
• Page 1 of 1
Who is online
Users browsing this forum: Google [Bot] and 7 guests