Reviewing the NS Inspiral S2 paper

1

• Reviewing the NS Inspiral S2 paper
• Inspiral review committee
• V. Kalogera, W. Kells, A. Weinstein (chair), A.
  Wiseman

2
What we are reviewing

• We have a rather mature first draft of a BNS S2
  paper Upper limit on the coalescence rate of
  Galactic and extragalactic binary neutron stars
  established from LIGO observations, 3/11/04
• There is extensive documentation of the details,
  in the e-log notebook
• The Inspiral Analysis Group has plans for more
  papers in the coming months
• (i) S2 Binary neutron star paper (present at
  March LSC)
• (ii) S2 Binary black hole search (present at
  June LSC)
• (iii) S3 Binary neutron star incl GEO (present at
  June LSC)
• (iv) S2/S3 MACHO binaries (present at Aug LSC)
• (v) S2 LIGO-TAMA search (present at Aug LSC)
• (vi) S3 Binary black hole incl GEO (present at
  Aug LSC)

3
Some things to look closely at

• The BIG PICTURE is the IAG addressing the right
  (astrophysical) questions? Is it covering all the
  questions it can/should? Is it using appropriate
  approaches? Is it organizing the papers and
  analysis sub-groups sensibly?
• Astrophysical motivations Are they adequately
  articulated in the paper? Are there flaws or
  concerns in the argument which should be
  addressed?
• Astrophysical models Are the templates modeled
  correctly? Are the bank tiled sensibly?
• Model dependence effect of spin higher order
  terms astrophysical effects such as orbital
  decay?

4
things to look closely at (2)

• Observation time is it well understood? Effect
  of chunks, segments, chunk edges, overlaps,
  etc. Data handling any data drop-outs due to
  bombed jobs? Due to incorrect overlapping?
  Forgotton cuts? What assurances / tests do we
  have?
• Data conditioning, filtering, line removal
  Sensitivity of result to data conditioning.Effec
  t of spectral features and non-stationarity on
  the result.
• Data quality Science segments, quality flags,
  vetoes. Granularity of PSD, calibration
  calculations. What assurances / tests that these
  are understood and handled correctly?
• Safety of vetoes and cuts assurances that no
  loud GWs have vetoed themselves.

5
things to look closely at (3)

• Hardware injections understand what they were
  used for and the checks / assurances they do and
  do not provide.
• Calibration how well is it understood,
  quantitatively? Checks assurances.
• Analysis pipeline and event tuning review of
  procedure, examination of code, searching for
  flaws, bugs, mechanisms for data to be lost or
  double-counted. Sensitivity to different tunings?

6
things to look closely at (4)

• Single-IFO Cuts SNR, chisq. Tuning.
• Coherent (multi-detector) cuts dt, matching of
  masses and D_eff. Handling of H1/H2 vs L1.
• Efficiencies, Monte Carlo simulations - review of
  procedure, examination of code, searching for
  flaws, bugs
• Calculation of source reach. Model dependence
  how well are source populations modeled?

7
things to look closely at (5)

• Background estimation. Statistics, cut
  dependence.
• Handling and evaluation of loudest triggers.
• Statistical analysis for upper limit.
• The paper clear, readable, accurate, complete,
  succinct?

8
The review process

• Over the next 2-3 weeks, the committee will
  examine all aspects of the analysis and the
  paper, formulate questions and request responses,
  and prepare a summary report
• Expect to meet a couple of times without IAG, and
  maybe once or twice with IAG chairs and members
• We welcome advice, comments, suggestions from all
  LSC members ajw_at_caltech.edu .