Offline multi ifo methods paper

Offline multi ifo methods paper organisation page

Personnel

• Marton Tapai
• Gareth Davies
• Thomas Dent
• Ian Harry
• Alex Nitz

Methods to demonstrate

• reduced SNR threshold
• clustering single triggers
• multi-dimensional background smoothing - in appendix: asssec:sngl_fitting
• multi-ifo coincidence
• multi-ifo ranking statistic(s)
• FAR combination

Analyses for testing

• Gaussian noise runs:
  • 3 detector 2ogc stat
    • CIT: /home/gareth.davies/analyses/pycbc/multiifodev/fake_o2/a22/HLV/output_3/
  • 2 detector 2ogc stat
    • CIT: /home/gareth.davies/analyses/pycbc/multiifodev/fake_o2/a22/HL-o3stat/output/
• O2 data
  • 3 detector analysis of chunk a20
    • CIT: /home/gareth.davies/analyses/pycbc/multiifodev/o2_comparison/a20_HLV/output_3/
  • compare this to the cvmfs analysis (as used in GWTC-1)

Need to consider:

• Injection sets (current sets are too close for fake design sensitivity searches) - current ideas:

  • Uniform in chirp distance between 5000 and 600000
  • BNS: mass 1 - 2.5, spin 0-0.4, uniform in component mass BNSSTT2
  • BBH: mass 2.5 - 50, spin 0-0.998, uniform in log mass. Spin aligned BBHSEOBNRv4
  • NSBH: mass1 2.5 - 50, mass2 1 - 2.5, spin1 - 0-0.998, spin2 - 0.-04, uniform in log mass

• Using PSD variation in Gaussian noise runs doesn't really make sense

Paper Writing

https://www.overleaf.com/23453475pcfkmqbrvwbj

As of 2020-01-29, Gareth thinks that the paper is in a good form and wants final checks

• Tom to look at the part where we discuss the choice of distributions of injections
• Alexander Nitz to check over the PTA consistency section -particularly the distribution of amplitude ratio for noise coincidences
• check/improve VT calculation definition
• chirp distance definition
• acknowledgements - CIT?
• Citation for BNSSTT2

2020-02-05, the paper is uploaded to DCC and ready to push the review button once we have a reviewer, and the following is todo after this point:

• sine-gaussian veto