Page 1 of 1

STEADY: convergence problems

PostPosted: Wed Sep 28, 2011 3:06 pm
by qlzxmu
Hello,
When i run the mod,there is a problem as follow:
Residuals of the static equations:

Equation number 1 : 0
....
Equation number 14 : 0
Equation number 15 : 0
Equation number 16 : -0.12337
... 0
Equation number 28 : 0
Equation number 29 : 0
Equation number 30 : -0.0035084
..
Equation number 36 : 0
SOLVE: maxit has been reached
??? Error using ==> steady_ at 132
STEADY: convergence problems

Is this means there is too large residuals in equation 16?
But when i directly put the initial value into equation 16, i found that the residual is close to 0.
What's wrong?Hope for you help.
Thank you!

Re: STEADY: convergence problems

PostPosted: Wed Oct 12, 2011 4:41 pm
by SébastienVillemot
The residuals that are printed by Dynare correspond to the initial values that you gave. So there is a discrepancy between what Dynare computes for equation 16 and what you get by hand. Either you made a mistake in your computation, or the equation you entered is not what you had in mind.

Then the steady state computation fails because the non-linear solver does not converge. Try another value for the "solve_algo" option of "steady". I had some success with "solve_algo=0".