STEADY: Convergence problems

Dynare help (legacy posts)
1

Dear all,

I am working on a deterministic model (OLG and frictions in the labor market).

When I try to simulate the model with Dynare, I get the following message:
SOLVE: maxit has been reached
??? Error using ==> steady_ at 27
STEADY: convergence problems

Error in ==> steady at 7
steady_;

Error in ==> ff_middle at 212
steady(0);

Error in ==> dynare at 26
evalin(‘base’,fname) ;

I computed the steady-states using Matlab and Mathematica.
By using the command resid (1), it seems that for two equations (the production function and the profit function) the residulal is not zero, even if I checked that the RHS is equal to the LHS.

Do you have any idea?

Thanks in advance,

Gabriele
ff_middle.mod (3.95 KB)

2

You have f and F as endogenous variables. You shouldn’t rely only on upper/lower case to distinguish two variables.

Dynare version 4 computes it correctly by the way, but it is still not a good modeling practice as too confusion prone.

Best

Michel

3

Thanks.

Changing the notation the model works.

4

Dear all,
I have somehow similar problem with convergence here.
I use fcsolve (written by F. Collard?) to solve for s.s value but whenever I put these s.s value into Dynare, the result is like this:

SOLVE: maxit has been reached
??? Error using ==> steady_
STEADY: convergence problems

Error in ==> steady at 7
steady_;

Error in ==> HouseBCCAC at 180
steady(0);

Error in ==> dynare at 26
evalin(‘base’,fname) ;

Any one has any idea?

Many thanks.

5

Hi DV,

This means that the steady state given by fcsolve (I do not know this routine) is not the steady state of the model you declared on the mod file. You may have a typo in your mod file (in the model definition or in the calibration of the model) or a typo in the equations used by fcsolve. Dynare sees that the proposed steady state is not a steady state and triggers a newton like algorithm to find the steady state. Unfortunately, our routine is unable to find the steady state of your model.

We have recently implemented an homotopy like routine for the computation of the steady states. We do not have a lot of experience with this routine, but you may give it a try. There is a description of this approach on the DynareWiki :

cepremap.cnrs.fr/DynareWiki/HomoTopy

Best,
Stéphane.