Dear all, 

I tried to google on the internet what the inverted roots problem is, as you recently told me about the ARMA-GARCH problem I had. 

Following up our discussion, the "best" model for my problem (under given circumstances) is AR(7)-GARCH(1,1)-t but the common inverse roots show up again. Can you please tell me what this means?

The Eviews estimation result is http://eubie.sweb.cz/gretl_forum/ar7.PNG

Thank you, 

Daniel