Zig-zag pattern in IRF from Tobin's Q adjustment cost

[derek]

I am working on a model with BGG accelerator and a banking sector. The capital production firm has a standard Tobin's Q quadratic adjustment cost for investment.

Somehow the IRF of the Q value flutuates in a Zig-zag manner, which causes all other variables in IRF to flutuate in zig-zag as well.

Has anyone come across this problem before and is there a simple reason behind it? Or the problem is more deeply rooted in my model equations or the parameter values?

Thanks in advance!
Derek

[jpfeifer]

This means you have a complex root that should not be there. It typically arises because of some timing error with respect to predetermined variables or the Euler equations. It could also be a matter of parameters but this is less likely.