Page 1 of 1

error "lnsrch1 at 53", Newton direction

PostPosted: Wed May 16, 2012 11:31 am
by Mahar
Hello,

This error message appears when i compute my code:
"??? Error using ==> lnsrch1 at 53
Some element of Newton direction isn't finite."

What does is it mean?
I changed a lot of initial values and parameters values, it isn't better.
I am almost sure the equations of the model are the right ones, and expressed in the right form.

Many thanks

Mahar

Re: error "lnsrch1 at 53", Newton direction

PostPosted: Fri Jun 08, 2012 11:45 am
by jpfeifer
Your starting values are so poor that Dynare cannot find the steady state. Try computing the steady state by pencil and paper and provide them to the program.

Re: error "lnsrch1 at 53", Newton direction

PostPosted: Sun Jun 10, 2012 8:31 am
by Mahar
Thank you for your answer, i'm going to find the steady state values by pencil and paper.

Regards

Re: error "lnsrch1 at 53", Newton direction

PostPosted: Fri Aug 02, 2013 5:18 pm
by selima
excuse me for my evident question , but what do you mean by computing steady state by pencil and paper ? is it not considering the rime subscript and putting the schock=0?
thank you for the answer.

Re: error "lnsrch1 at 53", Newton direction

PostPosted: Thu Aug 15, 2013 10:57 am
by jpfeifer
It means dropping time subscripts and setting shocks to 0 and then computing the steady state by hand on a piece of paper.