Dynare forums

Hi Professor Pfeifer,

I am trying to run a Bayesian VAR model, and I have followed the instructions in the paper "BVAR models 'a la Sims' in Dynare" to write my code.

Basically, the code I'm running is

"
var dx dy
varobs dx dy

bvar_irf(8,'Cholesky')

"

However, when I run this code, I receive the error message that although 2 variables are defined, no equations are defined.
As I followed the exact notation as in the paper listed above, I'm wondering what might be missing from this code.

Thank you so much for your time and help!

Please provide the full files to replicate the issue

Hi Professor Pfeiffer,

This is the code I used:

Code: Select all

var d_unemployment d_inflation d_depositrate d_lnusdtry;
varobs d_unemployment d_inflation d_depositrate d_lnusdtry;

bvar_irf(4,'Cholesky');
bvar_irf(50,'Cholesky');


I'm getting the following error message:

Code: Select all

Starting Dynare (version 4.4.3).
Starting preprocessing of the model file ...
ERROR: There are 0 equations but 4 endogenous variables!

Error using dynare (line 174)
DYNARE: preprocessing failed

I'm having issues attaching my dataset, but basically it is a four-variable dataset with 115 observations for each variable.

As I am running a BVAR and not a DSGE model, I thought that I did not need to specify a model and followed the same code as in "BVAR a la Sims", which consists of these lines of code, and not a model specification.

I would really appreciate it if you could assist me with this issue. Thank you so much for your time and help!

Isn't there a

Code: Select all

bvar_density

statement missing as in
https://github.com/DynareTeam/dynare/blob/master/tests/bvar_a_la_sims/bvar_standalone.mod

Regarding data-files, you should always upload everything in one zip-file

Thank you so much for your help!

I had one more question - is there a way to generate FEVDs for Bayesian VARs through Dynare?

Thank you again for your time Professor Pfeiffer

No, I am not aware of any functionality that would allow you to do this automatically.