Johansen Cointegration test on strend stationary series

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Bandleader
Posts: 2
Joined: Tue Jun 12, 2012 6:28 am

Johansen Cointegration test on strend stationary series

Postby Bandleader » Tue Jun 12, 2012 6:44 am

Hi all,

my question in concerning the Johansen cointegration test.

If I test 2 series for stationarity and they turn out to be trend-stationary, does it make sense to test them for cointegration?
I think if they have a time(deterministic)-trend, they are not cointegrated but just have a common trend which has nothing to do with cointegration.
And if so, what does option number 4 in the Johansen-test mean (EViews Users Guide II, pp. 686-688).
I think I just did not really understand which option to chose for what kind of series. I found no hint on that in this forum...

Thanks for a short answer!

hugo_eviews
Posts: 6
Joined: Tue Aug 29, 2017 11:45 am

Re: Johansen Cointegration test on strend stationary series

Postby hugo_eviews » Wed Aug 30, 2017 3:14 pm

Did you find the answer? I'm also puzzled by the use of cointegration methods on trend-stationary data. My understanding is that 2 variables can be cointegrated if they share a common stochastic trend, but is it also the case if they share a deterministic trend?

Thanks.


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 15 guests