Confusion of Wihout Steady State

Steady state
1

Hi everyone!
I have a confusion with my code.
When I finished the code, it can provide me a useful result.
But one thing is strange that when I copy the .mod file and paste in a new file, it cannot provide me the result anymore. After I restart matlab, the orginal code also does not work.

Here is the information offered by dynare:

Error using print_info (line 32)
The steady state has NaNs or Inf.

Error steady (line 102)
print_info(info,options_.noprint, options_);

Error no400.driver (line 518)
steady;

Error dynare (line 293)
evalin(‘base’,[fname ‘.driver’]) ;

Here is the code:
no400.mod (7.7 KB)

Any suggestions for me to make the code running?
Best wishes to the forum.
Thanks for any advice!

2 
b_h = g_h+l_e+lamda;
b_f_star = g_f_star+l_e_star+lamda_f;

are incorrectly linearized. The lambda are constant terms that should drop out.

3

Thank you for your guidance!
I am considering how to add the LTV rate into the model, but it seems that the LTV rate will be elimated after linearized.
I have another question, I make a setting that output is equal to the natural rate of out put in steady state. Could you please tell me is it right? I am not sure but I don’t know how to verify it.
Best wishes to you!

4
  1. The steady state will typically drop out, because everything is in deviations from steady state.
  2. Yes, in most models output in steady state will correspond to natural output as price rigidities asymptotically vanish.
5

Thank you very much!
When I am reading the orginal code, I found that the code use “model(linear);” , but the code doesn’t declare variables of steady state. In my opinion, the variables in steady state are necessary in lineared model. Could you please tell me why the model doesn’t declare steady state varoables in “parameters;” ?
I guess the reason is that the steady state variables are useless in the calculation of the model. I think so but I am not sure about that.
Again, Thanks for your reply!

Here is the orginal code:
no30.mod (4.6 KB)

6

The variables in a linearized model all have steady state zero. There is nothing to declare. Sometimes you need the steady state of the original level variables in the linearization. That is not the case if all original equations are multiplicative, which seems to be the case in the model you posted. Here, all original steady states drop.