Question about AR Processes

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Ynwe
Posts: 22
Joined: Tue Dec 08, 2015 4:39 am

Question about AR Processes

Postby Ynwe » Thu Dec 10, 2015 3:25 am

So I want to check if my equation has autocorrelation. So I type in AR(1) into the equation. But it also gives me a SIQMASQ variable out. Why does it do this? I remember in the past it didn't do this also the method is changed to a Method: ARMA Maximum Likelihood (OPG - BHHH)


Can't I just use my OLS Regression and add an AR term with it?

startz
Non-normality and collinearity are NOT problems!
Posts: 3797
Joined: Wed Sep 17, 2008 2:25 pm

Re: Question about AR Processes

Postby startz » Thu Dec 10, 2015 7:22 am

SIGMASQ is the estimated variance of the innovations to the error process.

Ynwe
Posts: 22
Joined: Tue Dec 08, 2015 4:39 am

Re: Question about AR Processes

Postby Ynwe » Thu Dec 10, 2015 7:55 am

is this a new feature in eviews 9? can remember seeing it in eviews 8

startz
Non-normality and collinearity are NOT problems!
Posts: 3797
Joined: Wed Sep 17, 2008 2:25 pm

Re: Question about AR Processes

Postby startz » Thu Dec 10, 2015 7:59 am

Yes, it's new.

guiri
Posts: 5
Joined: Mon Dec 21, 2015 5:25 am

Re: Question about AR Processes

Postby guiri » Mon Dec 21, 2015 7:57 am

please , i am working on prices indexes and i am trying to do the Ljung Box and ARCH tests, i have tetsted them with EVIEWS , i have written the steps to test it and the interpretation. I just want to know if my demarche is correct ?? because i should repeat this demarche with other 10 indexs! thanks a lot
Attachments
Ljung Box ,ARCH.docx
(114.84 KiB) Downloaded 522 times


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 2 guests