On Sun, 31 Oct 2010, Eloy A. Fisher Hogan wrote:
Hello everyone, I trying to migrate from Eviews to GRETL so
perhaps this is noob question - how can we state recursive
processes in GRETL?
I generate the nulldata 200 and create a random et using
randgen, I try to use the ARMA functionality but it doesn't
recognize the 0.8 coefficient, when I try to estimate yt I run
into problems. In E-views the code is:
smpl @first @last
series e=nrnd
smpl @first @first
series y = 0
smpl @first+1 @last
series y = .8*y(-1)+e
It's very similar in gretl:
nulldata 200
series e = normal()
series y = 0
y = .8*y(-1) + e
Alternatively, there's the filter() function.
Allin Cottrell