Need Help on Initial Values Setting

This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location where you will have to reset your password.
Forum rules
This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location (https://forum.dynare.org) where you will have to reset your password.

Need Help on Initial Values Setting

Postby Chazz » Mon Jan 31, 2011 4:06 pm

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.
Attachments
data.xls
(3.92 KiB) Downloaded 100 times
buildata.m
(1.94 KiB) Downloaded 108 times
model.mod
(15.78 KiB) Downloaded 86 times
Chazz
 
Posts: 2
Joined: Mon Jan 31, 2011 1:04 am

Re: Need Help on Initial Values Setting

Postby Chazz » Mon Jan 31, 2011 4:12 pm

Oups! I forgot to say that I had to change the "data" file extension to .xls because .csv cannot be uploaded.
Chazz
 
Posts: 2
Joined: Mon Jan 31, 2011 1:04 am

Re: Need Help on Initial Values Setting

Postby SébastienVillemot » Mon Feb 07, 2011 11:20 am

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.
Sébastien Villemot
Economist at OFCE – Sciences Po
SébastienVillemot
 
Posts: 706
Joined: Fri Dec 07, 2007 2:29 pm
Location: Paris, France


Return to Dynare help

Who is online

Users browsing this forum: No registered users and 7 guests