Page 1 of 1

Please help me out! about the QZ decomposition

PostPosted: Mon Dec 24, 2012 12:28 pm
by leelane
when I ran the .mod file, the error came out:

Error using print_info (line 36)
The generalized Schur (QZ) decomposition failed. For more information, see the documentation for
Lapack function dgges: info=19, n=19

Error in check (line 51)
print_info(info, options_.noprint);

Error in ngdptarget (line 242)
check;

Error in dynare (line 120)
evalin('base',fname) ;

I guess it related about the parametrition. I check almost every parameter and the calibration, and I can't find where the problem is. Any tip will be helpful. Thanks a lot.Here is code.

Re: Please help me out! about the QZ decomposition

PostPosted: Tue Dec 25, 2012 2:28 pm
by leelane
In fact, I have fixed the problem mentioned above.However, another bug emergd like this:
There are 9 eigenvalue(s) larger than 1 in modulus
for 10 forward-looking variable(s)

The rank conditions ISN'T verified!


Then I diagnose the model, using command " model_diagnostics(M_,options_,oo_)", and no error happened. This meaned no colinearity existed in the model, so I don't
know how to solve the problem. Here is the corrected code.

Re: Please help me out! about the QZ decomposition

PostPosted: Sat Dec 29, 2012 11:01 am
by jpfeifer
As always in those cases, there is a mistake with the timing somewhere.