The steady state contains NaN or Inf- Error in model or guesses?

Steady state
1

Hello,

I am new to dynare and wanted to use it for a little project. I set up a DSGE model to analyze fiscal policy measures (Taxes, subsidies) on the welfare of homeowners and renters under a shock to the real interest rate.
However, my code wasn’t working at all and I got “Impossible to find the steady state. Either the model doesn’t have a steady state, there are an infinity of steady states, or the guess values are too far from the solution”. Then I tried to go back to the basics of the model so without any of the extensions but still get “The steady state contains NaN or Inf” (see attached) .
Can anyone please tell me whether its the model or the guesses? ttt.mod (1.2 KB)

Thanks so much in advance!

2

Your initial values cannot be right. You set everything to 0, but

A/(1-exp(l))

then implies

A/(1-exp(0))=A/(1-1)=A/0=Inf
3

Dear Prof. Pfeifer,

thanks for the reply. I checked many different guesses but neither of them reduces the error. However, it changed to “Impossible to find the steady state. Either the model doesn’t have a steady state, there are an infinity of steady states, or the guess values are too far from the solution”.

So could this also mean the model has an issue?
Thanks in advance.

4

Please provide the most recent version.

5

Hello Prof. Pfeifer,

here the new version.
ttt.mod (2.0 KB)
Thanks,
Vanessa

6

Looking at the residuals after a call to steady, I get

Residuals of the static equations:

Equation number 1 : 0
Equation number 2 : 4.5613e-06
Equation number 3 : -3.305e-07
Equation number 4 : -1.1391e-06
Equation number 5 : 0
Equation number 6 : 0
Equation number 7 : 0
Equation number 8 : 3.5124e-07
Equation number 9 : 0
Equation number 10 : 1.8408e-06
Equation number 11 : 0
Equation number 12 : 4.9887e-07
Equation number 13 : -5.3015e-07
Equation number 14 : -1.3915e-06
Equation number 15 : -37.4346
Equation number 16 : 1.0098e-06

So there is something weird with equation 15. You should check whether that equation is really correct. In general, I recommend doing an exp()-substitution only after the model works. It is too error prone. One missing/wrong substitution will make finding the steady state impossible.

7

Thank You very much!
I will check this government budget constraint again.

I thought those exp() are needed for Dynare to do the log linearization. Is it possible to insert the model without those?

Thanks.

8

See

9

I tried to solve the steady state manually but end up with one equation and two endogenous variables (unknowns) H_o and H_r. So seems like one equation is redundant and another one missing.

Thanks for the help!

10

So there must be a mistake somewhere. Recheck everything.