Page 1 of 1

Error using ==> dynare_solve at 60

PostPosted: Fri Jun 16, 2017 1:02 am
by cassio
Good evening.

I'm having trouble finding the steady state of a DSGE model with banks. The following message always appears:

??? Error using ==> dynare_solve at 60
An element of the Jacobian is not finite or NaN.

I am a beginner and would like to know if there are any suggestions?

Thank you

Re: Error using ==> dynare_solve at 60

PostPosted: Fri Jun 16, 2017 7:26 am
by jpfeifer
First of all, only do the exp()-substitution once the model runs. Premature substitutions result in cumbersome code that is hard to debug. For example
Code: Select all
exp(A)      = rhoa*exp(A(-1)) + ee_a;                                                                                //  // (31)
exp(j)      = rhob*exp(j(-1)) + ee_j;                                                                                //  // (32)

has a steady state for e.g. exp(A) of 0, implying that A must be -Inf.

Also
Warning: Some of the parameters have no value (tal, mu, kappa, rhoc) 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...