Page 1 of 1

Need Help on Initial Values Setting

PostPosted: Mon Jan 31, 2011 4:06 pm
by Chazz
Hello,
I am trying to compute the steady state for the file attached below (with the data for estimation).
When I run the code, I get an error message indicating that the initial values are incompatible with some of the steady state equations.
I am also worried about the huge value found for the log posterior. Does the problem has anything to do with this?

Initial value of the log posterior (or likelihood): -2762594.3573

Max Line search Directional First-order
Iter F-count f(x) constraint steplength derivative optimality Procedure
0 43 2.76259e+006 -0.01
STEADY: numerical initial values incompatible with the following equations
2 3 34 47 48


If anybody has tips about the appropriate choice for initial values, please, keep me posted.

Thank you in advance,

Best Regards.

Re: Need Help on Initial Values Setting

PostPosted: Mon Jan 31, 2011 4:12 pm
by Chazz
Oups! I forgot to say that I had to change the "data" file extension to .xls because .csv cannot be uploaded.

Re: Need Help on Initial Values Setting

PostPosted: Mon Feb 07, 2011 11:20 am
by SébastienVillemot
This error means that some parameters are either not set, or are set to problematic values (for example a zero which is then used inside a log, or a negative value inside a non-integer power).

The problem occurs on your 2nd equation, on the 3rd, the 34th...

You need to manually investigate all the above equations to find where the numeric issue is.