(all) Smoothed variables missing

This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location where you will have to reset your password.
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.

(all) Smoothed variables missing

Postby lombard » Tue May 05, 2009 7:22 am

Hi,

I've estimated a rather large model (316 variables). While the mode finding and the MCMC seemed ok, Dynare did not return the SmoothedVariables in oo_.

I've run a similar model with the same set of options and I got all the expected output in oo_. The only major difference between the two models is that the one fully working is linear (it is the linearized version of the one that doesn't return all the output and has a simple rule instead of a full set of FOC of the policymaker).

Does anybody have a clue about when Dynare would not produce SmoothedVariables? (by the way, the model has no unit roots in either case) :?

Best regards

Gianni
lombard
 
Posts: 132
Joined: Wed Feb 02, 2005 12:36 pm
Location: Basel

Re: (all) Smoothed variables missing

Postby MichelJuillard » Tue May 05, 2009 8:18 am

Hi Gianni,

which version of Dynare are you running?
Did you have a list of variables after the estimation comand ? This is recommanded for large models
If not, how did you answer the question about which endogenous variables to include in the analysis?
If you use mh_replic=0,load_mh_file, is the problem reproducible?

If that the case, please, send me the files.

Best

Michel
MichelJuillard
 
Posts: 680
Joined: Thu Nov 18, 2004 10:51 am

Re: (all) Smoothed variables missing

Postby lombard » Tue May 05, 2009 8:53 am

Hi Michel,

This is the set of options I used
Code: Select all
options_.noprint=1;
options_.steadystate_flag=0;
options_.nograph=1;
options_.initial_estimation_checks=1;
options_.order=1;
options_.nomoments=1;
options_.bayesian_irf=1;
options_.load_mh_file=1;
options_.nocorr=1;
options_.irf=50;
options_.useAIM=0;
options_.prior_trunc=0;


I tried also w/o bayesian_irf and w/o load_mh_file.

I used a list of endogenous variables at the end of the estimation command.

Code: Select all
 estimation(nobs=103,nograph,datafile=nawmdata_long,presample=19,mode_compute=0,mode_file=nawm_fin_optimal_mode,  mh_replic=0,mh_nblocks=1) Y,C,INVS,X,IM,PII_Y,PII,PII_IM,Employ,PII_W1,Rn,S;


I'm going to sed you the folder with the model via e-mail.

Thanks and regards

Gianni
lombard
 
Posts: 132
Joined: Wed Feb 02, 2005 12:36 pm
Location: Basel


Return to Dynare help

Who is online

Users browsing this forum: No registered users and 13 guests