Catalog selection is always a problem in Seismicity. There are many local, regional
and teleseismic catalogs. My tendency is to always choose the local catalog
if it is available or the catalog which contains the best quality and quantity of
data. If I were a predictor, I think I would say in my prediction that evaluation
should be made using XXX catalog - for example for an event in Northern California
I might want evaluation to be done using the NCSN catalog, whereas for the Central
Atlantic PDE would be better. For Europe, probably choose ISC. In some cases,
however local catalogs are not the best choice. It all depends on the quality
of the data.
The NOA catalog is available on line at:
http://www.gein.noa.gr/English/home-en.html

Clearly, as you point out, evaluations can differ depending on the catalog against
which the prediction is compared.