All the contributions to the meeting address the need, for large mass Higgs-like resonanes, of applying a correction factor to the signal-only NLO model in order to account for the missing interference between signal and background in the analyses.
The various studies presented by CMS aim at calculating, at LO, a correction factor based on the comparison of signal-only (S) simulations (Madgraph) to signal+background (SBI) ones (Madgraph, Phantom).
The signal+interference (SI) is obtained by calculating the difference between SBI and background B only. The B case is modeled either by generating SBI with a Higgs resonance at 126 GeV (with Madgraph or Phantom), or by removing any possible diagrams with higgses from the calculation(with Madgraph).
The recipe presented for the uncertainty is based on the envelope of the results, obtained with scale variations from 0.5 to 2 times the nominal scale value.
ATLAS calculations are based on REPOLO, starting from VBFNLO calculations, and the uncertainty provided is by now covering only for an approximation performed with the aim of simplifying the correction.
VBFNLO is actually able and provide mWW shapes for S, B, SBI at NLO level.
Open points concern:
* the modeling of the SBI without accounting for the low mass resonance
* the scale factor needed to match the B to SBI away from the peak, in the ZZ > llvv case
* the possibility of reproducting the calculation at NLO with VBFNLO, to be then compared to the obtained results at LO
* the difference between considering the interference at W level or at fermions level
* the treatment of SI at low masses, at least for WW, where the distribution becomes negative: needs to be understood how to apply the correction at low masses with respect to the pole mass, where the SI is negative
* compare the corrections obtained with fixed scale to the ones one would get using the momentum transfer in the t channel
* perform a closure test of the reweighting in different variables than the WW mass, in the LO studies
Action items are by now the following:
* CMS WW madgraph5 studies will be carried on with B model including the 126 GeV resonance (Yiwen Wen, Qiang Li)
* CMS WW Phantom studies will implement the closure tests as well (A. Massironi, P. Govoni)
* CMS WW Phantom studies will implement the correction with running scale (using the momentum transfer in the t channel)
* All CMS studies will provide the uncertainty band calculated when variying the scales
* VNFNLO authors will circulate their results on interference at NLO, to be cross-checked with the LO corrections (here is the [[http://www.itp.kit.edu/~vbfnloweb/wiki/doku.php?id=download:beta][link]] to the code)
* ATLAS will circulate the recipe they plan to apply to calculate the uncertainty relative to the technique (Rikard Sandstrom)
There are minutes attached to this event.
Show them.