Hi Klaas,
===== snip =====
I do remember that in some cases the eigenvariate of a time series that
was whitened/HP-filtered may have peaks at the very beginning and the
very end of the time series, but I have forgotten the reason for this.
Maybe someone else on the mailing list can help.
===== snip =====
if I remember correctly, this happens if one doesn't adjust for an
"effects of interest" F contrast. In this case, the effects of filtering
the baseline regressor with the HPF will not be removed from the time
series and result in the peaks at the very beginning and end of the time
series.
Best,
Volkmar
<https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=SPM;12a3bf1.1003>
|