Re: [Gretl-users] New plot for arma models
Am 27.03.2017 um 10:46 schrieb Riccardo (Jack) Lucchetti:
Yeah, that's correct. But then, you'd be missing another nice
feature of
this plot, that is its ability to spot overparametrisation graphically.
What is known as the "common factor" problem in arma estimation is
essentially, in the frequency domain, a problem of overfitting, with the
model trying to drive the spectrum through random peaks in the periodogram.
Well I just hope people aren't interpreting this plot (haven't upgraded
and actually looked at one yet, sorry) in this way, namely trying to fit
the isolated peaks in the periodogram...
cheers,
sven