Hi,

When estimating a DSGE using the option dsge_var in the estimation command, how to obtain the coefficient of the VAR form, ie the Matrix A in

Y=ZA+U such as in the theoritical reminder?

Best regards.

See content of `oo_.dsge_var`

.

Dear Stepan,

following up your suggestion of looking at `oo_.dsge_var`

, and referring to the coefficients estimates matrix located in `oo_.dsge_var.posterior_mode.PHI_tilde`

. Which is the constant, the first or last row?

My guess is that it should be the first one, but I want to be sure about this.

Thank you in advance!

Yes, the first row stores the coefficients of the constant term.

Sorry for comment this,

I find that size of

oo_.dsge_var.posterior_mode.PHI_tilde

is n*rho by n, where n is number of observed variables and rho comes from

dsge_varlag

So, seems there is not constant term

I attach my codes

ls.mod (5.4 KB)

newmfile.m (1.9 KB)

All your model variables have steady state 0. Thus, Dynare set `noconstant`

to 1. That explains why there is no constant in your DSGE-VAR. However, your data is not mean 0. Thus, your observation equations must be wrong.

Dear all:

Iām new in Dynare.

As previous discussion, the first row in `oo_.dsge_var.posterior_mode.PHI_tilde`

is the constant term. I am wondering what is the way that store the rest of the coefficients in the `PHI_tilde`

?

For example, I estimated a dsge-var model with 3 lags,

and the list of the observable variables is:

`varobs A B C ;`

Does the coefficients stored like the way below, or like what else ?

const. const. const.

A(-1) A(-1) A(-1)

B(-1) B(-1) B(-1)

C(-1) C(-1) C(-1)

A(-2) A(-2) A(-2)

B(-2) B(-2) B(-2)

C(-2) C(-2) C(-2)

A(-3) A(-3) A(-3)

B(-3) B(-3) B(-3)

C(-3) C(-3) C(-3)

Thank you in advance!

Yes, it should be first the lag for all variables along the rows, then the second lag and so on.