Lagging independant variable in ARMA

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

jriskg
Posts: 6
Joined: Thu Jan 27, 2011 10:32 am

Lagging independant variable in ARMA

Postby jriskg » Thu Jan 27, 2011 11:07 am

Hi there

To be honest i'm new to eviews and am having issues in only this regard!

i have a very simple issue that i need clarity on please. I am having issues with putting in varying lagged independant variable into the ARMA regression code (i.e. x lagged once, x lagged twice, x laggred thrice, etc). Do i add in extra AR variables (i.e. AR(1); AR(2); AR(3); etc) and in doing so the independant variable is lagged, or do i run an ARMA (1,1) and actually lag the independant variable while keeping AR(1) and MA(1)???

Its a very simple regression, with one dependant variable and one independant variable.

Please help me on the code for the regression editor?!!!

Many thanks

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13586
Joined: Tue Sep 16, 2008 5:38 pm

Re: Lagging independant variable in ARMA

Postby EViews Gareth » Thu Jan 27, 2011 11:24 am

Do you want to include a lagged dependent variable, or an AR term? They are different when you have independent variables.

To include a lagged dependent variable, just have your equation like this:

Code: Select all

Y C X Y(-1) Y(-2)
etc...

jriskg
Posts: 6
Joined: Thu Jan 27, 2011 10:32 am

Re: Lagging independant variable in ARMA

Postby jriskg » Thu Jan 27, 2011 11:32 am

essentially i want to have a lagged independant term with the ARMA process.

So i simply do that? As opposed to including AR(1), AR(2), etc

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

Re: Lagging independant variable in ARMA

Postby startz » Thu Jan 27, 2011 11:42 am

Code: Select all

ls y c ar(1) ar(2) ma(1) x(-4)

jriskg
Posts: 6
Joined: Thu Jan 27, 2011 10:32 am

Re: Lagging independant variable in ARMA

Postby jriskg » Thu Jan 27, 2011 11:57 am

I dont mean to sound silly but you confused me now.

If i'm running an ARMA (1,1) model and i want to lag the independant variable is the code:

Y C X(-1) X(-2) AR(1) MA(1)

or do i lag through AR in the sense:

Y C X AR(1) AR(2) MA(1)

Thank you for bearing with me

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

Re: Lagging independant variable in ARMA

Postby startz » Thu Jan 27, 2011 12:02 pm

I dont mean to sound silly but you confused me now.

If i'm running an ARMA (1,1) model and i want to lag the independant variable is the code:

Y C X(-1) X(-2) AR(1) MA(1)

or do i lag through AR in the sense:

Y C X AR(1) AR(2) MA(1)

Thank you for bearing with me
I missed the ARMA(1,1) point.
The AR and MA terms refer to how the lag of the residual is modelled. If you want to have an ARMA(1,1) with the current value of the independent variable X, write

Code: Select all

ls y c x ar(1) ma(1)
If you want to use the first lag of X, instead write

Code: Select all

ls y c x(-1) ar(1) ma(1)

jriskg
Posts: 6
Joined: Thu Jan 27, 2011 10:32 am

Re: Lagging independant variable in ARMA

Postby jriskg » Thu Jan 27, 2011 12:11 pm

thank you so so much


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests