Hi;
Im trying to run a roll regression from the existing object eq01 (object roll01) but Im getting a message error (see the attached files).
The message is : _RSES02_NEER_SEYCHELLES01 is not defined.
Are there any solutions ?
Thank you
Rolling regression error message
Moderators: EViews Gareth, EViews Moderator, EViews Esther
Rolling regression error message
- Attachments
-
- erpt_estim.wf1
- (333.62 KiB) Downloaded 291 times
Last edited by Aktar on Wed Feb 21, 2018 10:16 am, edited 1 time in total.
-
- Fe ddaethom, fe welon, fe amcangyfrifon
- Posts: 13323
- Joined: Tue Sep 16, 2008 5:38 pm
Re: Rolling regression error message
EViews does not have rolling regression built in.
Follow us on Twitter @IHSEViews
Re: Rolling regression error message
I added the add-in roll
-
- Fe ddaethom, fe welon, fe amcangyfrifon
- Posts: 13323
- Joined: Tue Sep 16, 2008 5:38 pm
Re: Rolling regression error message
I added the add-in, runned an ols model and saved the object (see eq01), created a new object "roll" and selected the equation object for the rolling regression (fixed window ; step size1 ; window size = 47; start and last dates are the sames as the sample).
I didnt get the message for almost all models (cpi_xx c neer_xx price_xx), but for some countries, the neer_xx variable seems to get the estimation to crash although the variables seems fully normal.
Im getting the same message whatever I use an existing object or an equation entered manually.
I didnt get the message for almost all models (cpi_xx c neer_xx price_xx), but for some countries, the neer_xx variable seems to get the estimation to crash although the variables seems fully normal.
Im getting the same message whatever I use an existing object or an equation entered manually.
-
- Fe ddaethom, fe welon, fe amcangyfrifon
- Posts: 13323
- Joined: Tue Sep 16, 2008 5:38 pm
Re: Rolling regression error message
It sounds like you're using the roll user object, not the rolling regression add-in? You might be better off using the add-in.
Follow us on Twitter @IHSEViews
Re: Rolling regression error message
Yes, but i get the same bugs see the enclosed files.
I don't understand: when i replaced ar(1) by the lagged variable (ie cpi_xx(-1)) i got the bug... with the add-and the object... Is there a problem of sample size?
I don't understand: when i replaced ar(1) by the lagged variable (ie cpi_xx(-1)) i got the bug... with the add-and the object... Is there a problem of sample size?
- Attachments
-
- Capture1.JPG (236.75 KiB) Viewed 9990 times
-
- Capture2.JPG (232.18 KiB) Viewed 9990 times
Re: Rolling regression error message
ok it seems that the bug is coming from the character number of the variable label... reducing it makes the add-in working :s
Thanks
Thanks
Who is online
Users browsing this forum: No registered users and 33 guests