LSC-Virgo Burst Analysis Working Group


Burst Group Home
ExtTrig Home
CBC Group Home
LSC, LIGO, Virgo


'How-to' docs
[2008, 2007, earlier]
Talks/Posters [pre-wiki]
Paper plans
White papers


Analysis projects
Old notebook [General, S2, S3, S4, S5]
Virgo workarea
External collabs


Main review page

S5 Qpipeline High-Frequency Review Telecon 14th of August 2008 11:00 Eastern

Agenda and minutes of the S5 Qpipeline High-Frequency Review Telecon for Thursday the 14th of August at 11:00 Eastern time.

Agenda and Contact Info


Action items
  1. update on astro waveforms and Michele's link
  2. H1H2 analysis update
  3. Veto bug
  4. Update on Post processing script tests
  5. AOB

Contact Info

  InterCall telecon service:

    USA (toll-free): 1-866-616-1738
    UK (toll-free): 0800 073 8914
    Italy (toll-free): 800-906-494
    Germany (toll-free): 0800-1014-907

    participant code: 251 288 9495#


1. Astro waveforms:

Michele isn't present, so we will wait for next week for this agenda item. Brennan say there are some problems with rescaling of waveforms.

2. H1H2 analysis.

Jackie - describes tuning of the analysis using first year S5 data from just H1 and H2 when no L1 data is present (from webpage

Jonah - asks about whether Q-pipeline spits out the correlated/correlated energy

Jackie - these are calculated from the output

Jonah - Correlated energy should be positive definite, but thinks it can come out negative due to discretisation. Negative values should maybe just be mapped to zero. Check this.

Jonah - Was null stream used?

Jackie - Going to compare with null strean on and off

Brennan - We will end up using the null stream

Jonah - be careful about statements made about upper limits given null stream variation due to calibration uncertainties, given that efficieny may be different over the frequency range. Check on injections.

Jonah - for gamma threshold fits to give a false alarm rate were you using timeslides?

Brennan + Jackie - no, because of H1/H2 correlations and computational constraints. should be fine as it is though

Jackie - final descision use correlated energy cut of 21.1, null stream cut, but no gamma cut i.e. corr power not used

Siong - Reason that the null stream cut doesn't help for triple coincidence?

Brennan - presence of L1 means that the NULL stream doesn't make difference to triple coincident results. Null stream can in fact be /slightly/ detrimental.

Jonah - reproduce plots just using injections with some calibration uncertainty built in ,i.e. different amplitude in H1 and H2, and see if still get same results

Brennan + Jonah - checking anything caused by calibration phase uncertainties (Brennan has done some of this.) How does this effect things? Not really going for an upper limit statement in the paper, only a no detection statement, so it might not require too much looking into.

Siong - just because we're not having an upper limit we should still care about false alarm/rejection probability.

Jonah - yes, we should make sure that the phase uncertainty doesn't mean we miss things

Brennan - we could limit the frequency range to ones below where the phase uncertainty matters. we could not use the null stream veto. or we do checks and quantify.

Siong - method is significantly different to triple coincident analysis that it might require a lot of explaining.

Jonah + Brennan - should be same as low frequency analysis so we can just refer to low-frequency paper.

3. Veto bug 

Brennan  - Added more check to analysis scripts - one veto bug found. visibly changes things in figure 9, but not in the others.

Matt - I'll check through these and add to my webpage

4. Post-processing script tests

Brennan - added more checks and descriptions to previously described scripts

Matt - I'll check through these things as well.

5. AOB

Siong - away for the next three weeks, but telecons should carry on


Brennan Hughley, Jonnah Kanner, Matt Pitkin, Ik Siong Heng, Jackie Villadsen