Page 1 of 1

Posterior_moments Help

PostPosted: Fri May 22, 2015 3:41 pm
by Nice
Hello I have estimated a medium scale DSGE model using Bayesian Estimation, however HPD intervals and deciles are not being calculated it seems because a lack of draws. I get the following messages in the command window after the estimation. How can I correct for this? Also based on the information below does it seem like my estimation was conducted successfully? Your feedback is very helpful and much appreciated. Thanks in advance.

Initial value of the log posterior (or likelihood): -3046.1971
Fval obtained by the minimization routine: 3046.197147
Log data density [Laplace approximation] is -3132.820189.
SOLVE: maxit has been reached
Estimation::mcmc: Number of mh files: 1440 per block.
Estimation::mcmc: Total number of generated files: 2880.
Estimation::mcmc: Total number of iterations: 5000000.
Estimation::mcmc: Current acceptance ratio per chain:
Chain 1: 30.1695%
Chain 2: 30.4104%
Estimation::mcmc: Total number of MH draws: 5000000.
Estimation::mcmc: Total number of generated MH files: 1440.
Estimation::mcmc: I'll use mh-files 8 to 1440.
Estimation::mcmc: In MH-file number 8 I'll start at line 689.
Estimation::mcmc: Finally I keep 4975000 draws.

ESTIMATION RESULTS
Log data density is -3124.321241.
posterior_moments: There are not enough draws computes to compute HPD Intervals. Skipping their computation.
posterior_moments: There are not enough draws computes to compute deciles. Skipping their computation.
Total computing time : 198h10m11s

Re: Posterior_moments Help

PostPosted: Fri May 22, 2015 4:03 pm
by Daniel Bendel
This issue has been solved in this post:
viewtopic.php?f=1&t=5886

Re: Posterior_moments Help

PostPosted: Fri May 22, 2015 7:22 pm
by Nice
Thank you very much!