Page 1 of 1

Help for Zero Bayesian Posterior Variance of parameters

PostPosted: Mon Apr 25, 2016 4:10 pm
by Jesse
Dear All,
I am a PhD candidate who is busy writing my thesis regarding the application of the DSGE model, however, I have encountered the following problem. when i run the identification test, Dynare shows the following:
Testing prior mean
Evaluating simulated moment uncertainty ... please wait
Doing 360 replicas of length 300 periods.
Simulated moment uncertainty ... done!

WARNING !!!
The rank of H (model) is deficient!

WARNING !!!
The rank of J (moments) is deficient!

==== Identification analysis completed ====

It does not say explicitly which parameters are Not identified. But still says deficient.

When I run the model without identification test,
I find all the posterior variances of the estimated parameters are zero, I am wondering what kind of mistakes I have been made. The attachments are the mod file and the dataset
I wish somebody could kindly give some help about this? I need to submit my thesis soon and my uni does not have people who are good at Dynare to help me. Thank you very much and looking forward to hearing from you.
Kind regards,
Jesse

Re: Help for Zero Bayesian Posterior Variance of parameters

PostPosted: Sun May 01, 2016 4:07 pm
by jpfeifer
When I run it, I get
The rank of J (moments) is deficient!

ya is not identified by J moments!
[dJ/d(ya)=0 for all J moments!]


Are you sure your observation equations are correct? They are not mean 0 while the data are.

Re: Help for Zero Bayesian Posterior Variance of parameters

PostPosted: Mon May 16, 2016 11:54 am
by Jesse
Dear Johannes,
Thank you very much for your help! it improves the model!
Kind regards,
Jesse