P-values in DFGLS

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Moderator, EViews Jason, EViews Matt

evropeec
Posts: 4
Joined: Tue Sep 11, 2012 6:47 am

P-values in DFGLS

Postby evropeec » Tue Sep 11, 2012 6:52 am

I am running a bunch of unit root test and storing the results in a vector using the "save" option. Unlike ADF output, however, this option does not store the p-values for the DFGLS test in the fourth element of the output vector, even though the p-vlaue is reported in the series window's unit root view. I'd really appreciate it if there's an easy way to work around this issue, without copying the p-value from individual unit root views. Thanks!

EViews Glenn
EViews Developer
Posts: 2682
Joined: Wed Oct 15, 2008 9:17 am

Re: P-values in DFGLS

Postby EViews Glenn » Tue Sep 11, 2012 1:33 pm

Which version of EViews are you using?
Last edited by EViews Gareth on Tue Sep 11, 2012 2:08 pm, edited 1 time in total.
Reason: Preserve Queen's English.

evropeec
Posts: 4
Joined: Tue Sep 11, 2012 6:47 am

Re: P-values in DFGLS

Postby evropeec » Wed Sep 12, 2012 6:18 am

I'm using Eviews 7.1; thanks.

EViews Glenn
EViews Developer
Posts: 2682
Joined: Wed Oct 15, 2008 9:17 am

Re: P-values in DFGLS

Postby EViews Glenn » Wed Sep 12, 2012 9:36 am

I don't see where the DFGLS p-values are displayed...
Last edited by EViews Glenn on Wed Sep 12, 2012 1:48 pm, edited 1 time in total.

evropeec
Posts: 4
Joined: Tue Sep 11, 2012 6:47 am

Re: P-values in DFGLS

Postby evropeec » Wed Sep 12, 2012 11:51 am

Sorry--that was due to the alternating use of tau and t-statistics in the notes that I had. Anyway, it would be great if the p-value could be added to output, including using the save option.
Thank you.

EViews Glenn
EViews Developer
Posts: 2682
Joined: Wed Oct 15, 2008 9:17 am

Re: P-values in DFGLS

Postby EViews Glenn » Wed Sep 12, 2012 1:48 pm

There is no p-value. Only the tablulated critical values. To the best of my knowledge, there have not been any published response surface approximations to the tabulated values, which is why there are no p-values.

evropeec
Posts: 4
Joined: Tue Sep 11, 2012 6:47 am

Re: P-values in DFGLS

Postby evropeec » Thu Sep 13, 2012 6:15 am

Thanks for clarifying this. It would be great if the critical values could be added to the output file, so that comparing test stats to them can be carried out without looking at the individual series' windows.

EViews Glenn
EViews Developer
Posts: 2682
Joined: Wed Oct 15, 2008 9:17 am

Re: P-values in DFGLS

Postby EViews Glenn » Fri Sep 14, 2012 3:44 pm

I'll make a note of the suggestion. Thanks.


Return to “Programming”

Who is online

Users browsing this forum: No registered users and 2 guests