Page 1 of 1

Less forward-looking variables than good eigenvalues

PostPosted: Tue Mar 15, 2016 7:44 am
by Palable
Hi Dynare expert,

I was modifying models and get the error message:

"There are 10 eigenvalue(s) larger than 1 in modulus for 9 forward-looking variable(s). The rank condition ISN'T verified!".

I checked the forums, most problems involve less eigenvalues larger than 1 but more forward-looking variables.

I played around with the code, one easy fix is to roll one of my equation forward. (for example, make exp(N)= exp(Ne) to exp(N(+1))= exp(Ne(+1))). But I am not sure this corresponds to correct model dynamics. I am wondering is there something fundamentally wrong?

Thanks.
P

Re: Less forward-looking variables than good eigenvalues

PostPosted: Tue Mar 15, 2016 8:01 am
by jpfeifer
Yes, usually there is something fundamentally wrong, typically the timing of some predetermined variables like the (capital) stock. There is one and only one correct timing dictated by economic intuition. You cannot simply shift some variables by one period.

In your case, it might for example have to do with the split between
Code: Select all
exp(N)      =   exp(Ne)+exp(Nn);

see http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=3564