I am using Dynare 4.3.2 to apply Bayesian Estimation to CTW (2010) model. There are puzzles in my results.
The posterior distributions are very narrow when I use estimation with mode_compute = 6, compared to priors. They are just like spikes.
Since I don't really know how can I solve?I look forward to any hints for this puzzle I encounter. Thank you!
Dynare forums
Forums for asking questions, posting comments and uploading examples related to Dynare
The posterior distributions are just like spikes
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.
2 posts
• Page 1 of 1
The posterior distributions are just like spikes
by liuhuazhi » Tue Jul 23, 2013 2:34 pm
- liuhuazhi
- Posts: 1
- Joined: Wed Apr 03, 2013 2:25 pm
Re: The posterior distributions are just like spikes
by jpfeifer » Wed Jul 24, 2013 10:11 pm
Usually that's a bad sign. Have you checked the jscale coming out from mode_compute=6? Often it is adjusted to provide a decent acceptance rate and only achieves this by making mh_jscale so small that the MH chain does not wander around in the posterior. If my conjecture is correct, try mode_compute=9. Moreover, try two chains (mh_nblocks=2) and look at the mode check plots and the (mode_check) convergence diagnostics.
- jpfeifer
- Posts: 6940
- Joined: Sun Feb 21, 2010 4:02 pm
- Location: Cologne, Germany
2 posts
• Page 1 of 1
Who is online
Users browsing this forum: No registered users and 7 guests