EViews.com

dear forum,

I am trying to run Pedroni's panel co-integration test. I am using Eviews 8.

My data is structured as panel. To start co-integration test, I checked Eviews 8 help . It says "select views/cointegration test/panel cointegration".

however, I don't see this command under views. I rather see quick/group statistics/johansen co-integration test. when I run johansen, it automatically takes me to "panel co-integration tests"

Am I on the right track? Is my data correctly structured as panel? I carefully defined it as panel by following the directions here: http://www.eviews.com/Forum_support/pan ... ucture.htm

Can you check to see if my data is correct panel format? I have ATTACHED it. I don't see any command of "panel co-integration test" in Eviews 8. Instead, Johansen co-integration leads to panel co-integration test.

Is johasen and panel co-integration test the same?

I really appreciate your help. Thanks a lot in advance.

lnp3
Mine.D.

Open up a group, then select View.

how can I open up a "group"? thanks a lot really.

M.D.

ok I was able to figure out. I created a new group: object/new object/group

after creating the group, I go to "view" within the group.. co-integration is listed under the group views menu ..

thanks.

I also figured out that it gives the same results if I do the other way.

without creating a group object.. I can write the list of variables under johanson co-integration test (quick/group statistics/johansen), and it automatically takes me to panel-cointegration..

best regards,
M.D.

I want to test Panel Pedroni Cintegration test using E-views. I just want to that, is there any possibility of adding "Common Time Dummies" in cointegrating equation in Eviews 8?

As I replied in an identical thread...

Time dummies are not built into EViews 8.

If you have a balanced panel then I believe that you can remove the common time effects yourselves prior to performing the testing. I believe that both Perron and Mark and Sul discuss this issue. I do vaguely recall, however, that these results implicitly assume a balanced panel.