matrix close to singular or badly scaled

[qzx]

Dear all,

I log-linearized my model myself and simulated it in dynare which gives me the moments and seemingly sensible impulse response functions. However, I get a warning that "matrix close to singular or badly scaled" somewhere. What does this suggests? Do I have to worry about it?

Thanks.

Best,

Zongxin

[qzx]

Dear all,

The problem is resolved. It is due to a nonstationary variable inclued.. No need to bother with the question anymore. Thanks.

Best,

ZX