PAS QA for B2G-15-001

Documentation

• CADI entry

Color code

• Answers in blue are considered final.
• Answers in orange require further communication between the questioner and the authors.
• Answers in red still need work from our side.

Comments/questions received before pre-approval

From Justin Pilot, 1st of April 2015

• L53 - this "FIXME" is a good idea. Especially for the PAS, you should make clear what are the differences between your theory model and the one used by ATLAS. If some results are directly comparable to ATLAS' results (or not), this should be explained so unfair comparisons are not made.

• Table 1 - caption says describes something for MC -- should be changed.

• For the PAS, I would avoid the term "Monte Carlo" in favor of "simulated events" or something similar.

• Section 3.2.3. It would be good to add some signal validation plots here -- maybe the MET distributions for various mass points and couplings? A comparison of the signal acceptance for each?

• Figure 2 is missing? (Related to your PDF problem perhaps?)

• L119 - is the HLT_IsoMu24_eta2p1 trigger fully efficient for your event selection? Do you have a plot of the turn-on curve?

• Paragraph at L161 -- do you have a "N-1" plot to motivate this cut, comparing signal and background?

• L170 - again here a plot of N-jets and N-btags, and m_T(W) comparing signal/background would be useful to motivate the selections

• L227 - does the isolation veto give you a shape uncertainty for the final results, or just a normalization component?

• (Fig 19 e.g.) It is sometimes hard to read the plots with all of the error bars on the histograms and the data points. Does the shaded band on the stack plots include all uncertainties? If so, I suggest removing the error bars on the stacked histograms.

• A question on the "mT(W) tail" nuisance parameter -- this is what is described about using the different shapes of the W+c, W+b, W+light jet distributions in Section 5, correct? Maybe you can add a short description in the list of systematics in section 5.2. Does this give a shape difference only in the tail of the distribution, or is it a shape difference in the entire distribution of mT(W).

• For the eventual public limit plots (and others), you should switch to the new style recommendations: https://ghm.web.cern.ch/ghm/plots/

• Figure 20 - Is it valid to show the theory for cross sections with a = 0.05, when the samples are generated with a = 0.1? Would the limits change if using signals with a = 0.05?

-- JeremyAndrea - 2015-04-01