Impulse Response in simple AR

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

strugglingeconomist
Posts: 3
Joined: Tue Apr 03, 2012 4:08 pm

Impulse Response in simple AR

Postby strugglingeconomist » Tue Apr 03, 2012 4:15 pm

Hello,

I am trying to find strucutral breaks in inflation series via a simple AR(6) process. I am not using VAR but would still like to do impluse response analysis. Is this in anyway possible in eviews?

I'd really really appreciate any help.

Thank you..

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Impulse Response in simple AR

Postby startz » Tue Apr 03, 2012 4:28 pm

I'm not sure what this has to do with a structural break, but if you estimate an AR model and choose View/ARMA structure you can see the impulse response.

strugglingeconomist
Posts: 3
Joined: Tue Apr 03, 2012 4:08 pm

Re: Impulse Response in simple AR

Postby strugglingeconomist » Sat Apr 07, 2012 2:42 pm

Hi!

Sorry for the late reply. I wasn't informed that you had replied to my query.

I did View/ARMA structure but the error msg said: "Equation does not have ARMA terms" (I figured because it had only AR terms.. :?: )

ps: the structural break comment was a miserably sad attempt to give you some background to what I am trying to do, which is: econometrics on inflation dynamics - so persistence, structural breaks and impulse responses on inflation series (if I can manage the last one)... Can I? Please please say yes.

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Impulse Response in simple AR

Postby startz » Sat Apr 07, 2012 3:51 pm

Just AR models work fine, but you have to use the AR psuedo-variable. For example, if you have an AR(2) model

Code: Select all

ls y c ar(1) ar(2)
Alternatively, you can set up a VAR with only one endogenous variable.

strugglingeconomist
Posts: 3
Joined: Tue Apr 03, 2012 4:08 pm

Re: Impulse Response in simple AR

Postby strugglingeconomist » Mon Apr 09, 2012 1:21 pm

I am not sure how you set up a VAR with only one variable. But I just did the method you described and it worked perfectly! Thank you so much

Though I didn't understand what was "1s" in the beginning...? o.O

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Impulse Response in simple AR

Postby startz » Mon Apr 09, 2012 2:34 pm

I am not sure how you set up a VAR with only one variable. But I just did the method you described and it worked perfectly! Thank you so much

Though I didn't understand what was "1s" in the beginning...? o.O
"LS" if used at the command line.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 1 guest