Problem with simulating ARMA(2,1)

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

kirillov
Posts: 4
Joined: Wed Feb 16, 2011 8:57 am

Problem with simulating ARMA(2,1)

Postby kirillov » Wed Feb 16, 2011 9:30 am

Hello!
Help me, please. I have been generating (with "program") time series, consisting of 2500 observations and its process is ARMA(2,1):

smpl @all
series e=nrnd

smpl @first @first
series x=1

smpl 2 @last
series x=1+0.5*x(-1)+0.1*x(-2)+e-1.2*e(-1)

EViews has been getting only first value of the series, other values are NA. What could I do?

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

Re: Problem with simulating ARMA(2,1)

Postby startz » Wed Feb 16, 2011 9:56 am

You have two lags of x, so you need to start the smpl at 3.

kirillov
Posts: 4
Joined: Wed Feb 16, 2011 8:57 am

Re: Problem with simulating ARMA(2,1)

Postby kirillov » Wed Feb 16, 2011 10:03 am

Thanks very much!


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests