icecube / i3mla

Public IceCube point-source analysis tools
Apache License 2.0
2 stars 1 forks source link

Better grl #47

Closed thejevans closed 3 years ago

jasonfan1997 commented 3 years ago

I think the idea of distributing events base on the grl rate is good, but if we don't change the corresponding time likelihood maybe it could cause a false positive. Also, I think the injected signal events should not weight based on the grl rate The original idea of disabling the grl filter is to stop filtering out injected signal events when the n_signal_observed is given so it should still be keep.