Seem to remember if an ar model has b1 close to 0, say 0.995, can we conclude unit root exists? If so why do we still bother with docket fuller test? Or I am wrong? Thanks
I think you still need to do the test. The raw figure gives you the inclination to test. Otherwise, you wouldn’t bother testing (if its not close to 1 already).
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.