Automatic selection for DF GLS test - 2

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

kiber_master
Posts: 94
Joined: Fri Sep 23, 2011 3:56 am

Automatic selection for DF GLS test - 2

Postby kiber_master » Wed Oct 17, 2012 2:24 am

Hello again!

Could you explaine, what does mean "t-statistic" criteria? for what should I set p-value for this criterion? I can't find info in Userguide.

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

Re: Automatic selection for DF GLS test - 2

Postby EViews Glenn » Wed Oct 17, 2012 5:04 pm

Test successively higher lag terms for statistical significance using a t-test with the specified p-value. Stop when there is a lag that is insignificant according to the test. It's a relatively new option that escaped documentation. It's a pretty standard approach that is discussed in all of the papers.

kiber_master
Posts: 94
Joined: Fri Sep 23, 2011 3:56 am

Re: Automatic selection for DF GLS test - 2

Postby kiber_master » Thu Nov 22, 2012 2:43 am

Sorry, but I didn't catch =(
Could you give a reference at least on one of the papers you mentioned?

And does this autoselection method correct sample using max lag, as other methods do?

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

Re: Automatic selection for DF GLS test - 2

Postby EViews Glenn » Mon Nov 26, 2012 11:51 am

Ng and Perron “Unit Root Tests in ARMA Models with
Data-Dependent Methods for the Selection of the Truncation Lag,” JASA, 1995.


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 2 guests