Dear all, Hello,
I would like to ask a question about heteroskedasticity test using Eviews 11.
Actually, after having the results of my estimations using fixed effect model for paned data (for 2 countries; period of 17 years) when i try to run the test of heteroskedasticity (diagno test) eviews dont allow me, and share as message : "cross-section effects with period GLS weights not allowed". I dont know why !
If any one can share any response, i would be grateful.
Think about it and save the residuals and create the test instead of relying on E-views to do it for me.
Think about it and save the residuals and create the test instead of relying on E-views to do it for me.
Dear Chuck,
Thank you for your response ! Please note that this is the first time that im using eviews ! And im new in econometrics.
Can you share more details related to your answer, i mean how can i save the residuales and what do you mean by creating the test ?
Thank you in advance
Fatema -
You said it says "cross-section effects with period GLS weights not allowed." Here's a guess: Could it possibly mean that Eviews saves GLS for when you are looking at longitudinal data, and what you need here is just WLS for the heteroscedasticity? Could you try saying you want WLS instead of GLS?
Just a guess.
Best wishes - Jim
Dear James,
Hello,
Thank you for your reply ! But i didnt understand your reponse, if you can share more details.
Thank you
I don't know Eviews, but I'm going by that quote: "cross-section effects with period GLS weights not allowed."
It sounds like Eviews won't allow your GLS weights because of the autocorrelation consideration. If there is a manual you can find online, or you have one, you could look to see if you have specified something incorrectly, such as the weights you are using, perhaps.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running