Storing Variables of Unit Root Test Output

For making suggestions and/or requests for new features you'd like added to EViews.

Moderators: EViews Gareth, EViews Moderator

thaigrrr
Posts: 9
Joined: Wed Feb 27, 2013 9:49 am

Storing Variables of Unit Root Test Output

Postby thaigrrr » Thu Mar 07, 2013 11:27 am

I'm writing a program to take a group of series through unit root test one-by-one. The program is testing all three uroot options--const, trend, and none--but the code is quite lengthy because the output table for each option is not in the same size. So, if I want to keep DW for instance, then I'd have to specify the table position differently for each option. From what I've read/found on the forum, there's no easy way around this (correct?). It'd be a nice option to store these output variables--besides C and resids--in something that can be easily recall.

Thank you :)

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13319
Joined: Tue Sep 16, 2008 5:38 pm

Re: Storing Variables of Unit Root Test Output

Postby EViews Gareth » Thu Mar 07, 2013 11:40 am

Actually, I believe the uroot command has an save= option that lets you specify the name of a matrix in which to store results. The results aren't labelled, but if you look at the output once, you should be able to work out what is what.

Code: Select all

gdp.uroot(save=myoutput)
Follow us on Twitter @IHSEViews

thaigrrr
Posts: 9
Joined: Wed Feb 27, 2013 9:49 am

Re: Storing Variables of Unit Root Test Output

Postby thaigrrr » Thu Mar 07, 2013 11:55 am

Hi Gareth,

You're right about the save= option. Unfortunately the information saved is quite limited, and not the ones that I need. (I think it's only saving obs, max lag length (but not the actual no. lag used), ADF t-statistic, and ADF p-value. I couldn't match up the rest.) Perhaps save= option can be expanded to include other output variables (if not all)?

Thank you :)

thaigrrr
Posts: 9
Joined: Wed Feb 27, 2013 9:49 am

Re: Storing Variables of Unit Root Test Output

Postby thaigrrr » Thu Mar 07, 2013 1:22 pm

Just to quickly add to my previous comment.

It looks like the lag length used is recorded with save= option. So, I assume that all the NAs I'm seeing now (and couldn't match previous) are most likely associated with other options that weren't used here.


Return to “Suggestions and Requests”

Who is online

Users browsing this forum: No registered users and 25 guests