-
Notifications
You must be signed in to change notification settings - Fork 356
Multiifo Ranking Statistic Development
Gareth Davies edited this page Jun 28, 2019
·
21 revisions
| Name of problem/issue | Status | Testing/Verification |
|---|---|---|
| Total noise rate for each type of coinc | Done | predicted rate vs timeslide rate |
| Relate total number of single triggers to thresholded fit | Not yet, set up plotting script | see distribution and fitting of below-threshold triggers |
| convert predicted noise rate into statistic | Done, exp_fit_sg_coinc_rate statistic in stat.py. See #2689 | coincident trigger scatter plot, and comparison histograms (LIGO login required). |
| Proper combination of different types of coinc via statistic value (not FAR/trials factor) | Work in progress, pycbc_multiifo_add_statmap see #2811 | compare unclustered event FAR assignments from trials factor vs adding histograms |
| Approximation of network sensitivity from sigma^2 | Not yet, include in coinc_rate module | plot sensitivity vs time /mass/etc. compare to standard plots |
| Add in network sensitivity into statistic | Not yet, into stat.py | Newstat vs previousstat as function of time/coinc type |
| Signal extent over time difference, include in signal rate part of statistic | In coinc_rate module | compare to noise trigger extent allowed time differences for noise and signal triggers |
