Some parameters have no value when using steady

Hello everybody,

I am trying to run this model, that worked before adding capital and capital adjustment costs and the government budget constraint.
I wrote the steady-state in the steady_state_model block. I also tried to write it in the “parameters” block, as you write in the Dynare Manual, or as # parameters as suggested in this forum.
I double-checked the steady-state, I linearized again the government budget constraint, but I am still missing something. It has been days since I am stuck at this point. The error is:

Warning: Some of the parameters have no value (y_ss, a_ss, inv_ss, pi_ss, r_ss,
n_ss, w_ss) when using steady. If these parameters are not initialized in a
steadystate file or a steady_state_model-block, Dynare may not be able to solve
the model…
STEADY: numerical initial values or parameters incompatible with the following equations

This is the government budget constraint. It has “Residuals: NaN”
I do not know where the error comes from: mistakes in linearizing, mistakes in the steady-state, or mistakes in writing the steady-state in the .mod file. Or numeric values are to be assigned to y_ss, a_ss, inv_ss, pi_ss, r_ss, n_ss, w_ss?
new.mod (3.3 KB)

Thanks to all of you who could help me.

See e.g.

Hi Claudia,

your “Steady state relationships” do not work, that is why you get NaN’s. If you run your model you see that most of them are NaN. One example, you declare those two under the parameter block:

pi_ss=r_ss*betta; r_ss=pi_ss/betta;

But for the first you need the second and vice-versa. In the parameter block you need to give Dynare numbers, this is why it does not work. As long as your ss-values have NaN, your equations featuring those will necessarily also have NaN’s.
If you have calculated the steady state already write the number directly in the parameters block as those are parameters. Hence, yes numeric values need to be assigned to those parameters (y_ss,…).

If you are unsure about your linearization, why don’t you try the non-linear model and let Dynare to the linearization for you?


Thank you so much @jpfeifer and @DoubleBass. That was really helpful, I solved the problems with the calibration of SS variables.
Since I do have indeterminacy issues now, I will also try running the non-linear model.
Thanks again