Blanchard Kahn Conditions not Satisfied

[paulinares]

Hello, I'm trying to estimate a model but I'm getting the following error:

Error in computing likelihood for initial parameter values
??? Error using ==> print_info at 39
Blanchard Kahn conditions are not satisfied: no
stable equilibrium

What is puzzling for me is that I use the command check, and the rank condition is verified
(There are 20 eigenvalue(s) larger than 1 in modulus for 20 forward-looking variable(s)), and
the policy functions are being calculated without a problem.

If the rank condition is verified why is there a problem with the BK conditions to compute the
likelihood?

I've tried to move my initial parameter values around as well, but nothing helps. I do have several
infinite eigenvalues, I thought this isn't an issue because the code must be using a QZ decomposition
but can this have something to do?

Thank you in advance for your help!
Paulina

[jpfeifer]

Did you put your parameter values as starting values in the estimated_params-block? Or only before the model block?

[paulinares]

Thanks so much for replying!

I only put the parameter values before the model block. In the estimated_params-block I have
the prior (parameter name, distribution, alpha, beta), but not the starting value. Does that make
a difference?

Thanks again!
Paulina

[jpfeifer]

Yes, because if you do not specify starting values, the posterior mean is used. This may explain your error with initial values for estimation.