High Persistence MA Model

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

diggetybo
Posts: 152
Joined: Mon Jun 23, 2014 12:04 am

High Persistence MA Model

Postby diggetybo » Tue Feb 24, 2015 12:39 pm

Hey,

I have a question on how eviews handles a high persistence moving average process in the error.

My dependent variable has this kind of process even after seasonal adjustment; refer to this graph of ARMA stucture correlogram:

arma structure.jpg
arma structure.jpg (34.46 KiB) Viewed 2745 times


According to the view ARMA structure my left hand side variable has a pronounced MA process well after 100 lags.

Can eviews compensate for this? I tried adding 100 ma terms, but eviews said: "Too many ARMA terms"

I'm using eviews 6, so I can't just type ma(1,100) like the latest version.

Please advise :)
Last edited by diggetybo on Tue Feb 24, 2015 1:27 pm, edited 1 time in total.

EViews Glenn
EViews Developer
Posts: 2616
Joined: Wed Oct 15, 2008 9:17 am

Re: High Persistence MA Model

Postby EViews Glenn » Tue Feb 24, 2015 1:05 pm

These days, persistent autocorrelations are typically modeled using fractionally integrated models which allow for separate handling of long-run persistence and short-run dynamics.
EViews 6 does not support fractional estimation, but EViews 9 (which is in beta testing for those who have EViews 8 ) offers ARFIMA estimation.

diggetybo
Posts: 152
Joined: Mon Jun 23, 2014 12:04 am

Re: High Persistence MA Model

Postby diggetybo » Tue Feb 24, 2015 1:29 pm

I see,

Thank you, I will look forward to eviews 9.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 4 guests