-- AmnonHarel - 03-Sep-2010

What dijet masses should be used in a normalization region for the dijet centrality ratio analysis?

Why go high?

  • reduce lever arm from normalization region to the region that's most relevant for setting limits
    • currently aiming at limits ~3TeV --> most relevant region is ~1.6-2.4TeV c3lever.png

  • reduce effect of (and hence uncertainty on) non-perturbative corrections
    • the low Mjj rise ends at around 0.6TeV. Above it the correction is 2%. At 400GeV it's 5%. At 200GeV 10%. ratio_pythia_MpiHadvsNoMpiHad.png (original .pdf image)

Why go low?

  • no point in discrimination power against lambdas that have been repeatedly and easily ruled out at the Tevatron - aim to discriminate against 2TeV and up --> Can probably start as high as 1.2-1.4TeV.

  • to have a larger signal region where we reproduce and support the Tevatron results on contact interactions
summary plots
Run I: Ecm=1.8TeV
CDF's 1996 angular distribution D0's 1998 angular distribution D0's 2000 multijet HT analysis
(R defined in terms of chi)  
CDF_dijet_angular_dist_PRL_77_5336_1996_money_plot.png D0_dijet_ang_dist_PRL_80_666_1998_money_plot.png D0_jet_HT_compositeness_PRDRC_62_031101_2000_money_plot.png
Run II: Ecm=1.96TeV
CDF's dijet cross-section D0's dijet cross-section D0's 2009 angular distributions
CDF_Mjj_PRD_79_112002_2009_fig1b.png /www-d0.fnal.gov/Run2Physics/WWW/results/final/QCD/Q10A/Q10AF01.jpeg www-d0.fnal.gov/Run2Physics/WWW/results/final/QCD/Q09C/Q09CF1.jpeg
    • the dijet mass plot indicates good coverage out to 1.1-1.2TeV
    • but the dijet angular distributions seem weak from around 0.7TeV.
    • in particular, excited quarks have been excluded only up to 870GeV. Here's what an 880GeV q* would look like in out analysis: 880qstar.png
though it's not the focus of the proposed paper, we insist on having an overlap with the Tevatron, even for q*s.

  • irrelevant to have a larger signal region where we probe QCD @ 7TeV - no - we don't use the normalization for the QCD part

Other considerations

  • higher statistics --> go to low end of a trigger, but there are several triggers to choose from
    • currently, our triggers turn on at bins #1 (156GeV), #5 (244GeV), and #9 (354GeV), so this simply leads us to prefer low (and wide) normalization regions

Preliminary discussion

Since the main motivations to go low cannot be quantified, we need to make a judgement call. * There's no real reason to go below 600GeV (bin #16) * Rather end the normalization region below ~800GeV (bin #20 starts at 788):
  • The Tevatron q* limits run out at 870GeV (bin #20 starts at 788). We can rule out such a q* from the ratio by looking at Mjj from 800-1000, indicating that the Tevatron didn't really cover this region.
  • See above how the dijet angular measurements started loosing precision around there (e.g. D0's Run II started wider binning at 900GeV, can't resolve low chi rise at 700-800)

--> Use 400-600 or 600-800GeV. But why only a narrow range, can gain from a wider one...

Time to look at the detailed plots from Jim

  • statistical errors go up if using only the very high bins, e.g., 600-800 (16-19). But ending in 19 is reasonable if starting low enough (13 = 489).
  • unless we push the normalization region above 800GeV, it's hard to significantly reduce the lever arm effect ("NLO-QCD") - will this change when Jim adds flat model to lever arm calculation?
  • as expected, all these are reasonable as far as the NP correction uncertainties go

Indeed, given the constraint of not normalizing with data above 800GeV, Jim's quantitative analysis yielded for the sum that this (13-19) is one of the best points.

Bottom line

Will use bins 13-19 (489-788GeV) as the normalization region.

  • 880qstar.png:
Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng 880qstar.png r1 manage 12.1 K 2010-09-05 - 20:37 AmnonHarel  
PNGpng CDF_Mjj_PRD_79_112002_2009_fig1b.png r1 manage 57.7 K 2010-09-03 - 11:51 AmnonHarel  
PNGpng CDF_dijet_angular_dist_PRL_77_5336_1996_money_plot.png r1 manage 86.9 K 2010-09-03 - 12:06 AmnonHarel  
PNGpng D0_dijet_ang_dist_PRL_80_666_1998_money_plot.png r1 manage 11.7 K 2010-09-03 - 10:34 AmnonHarel A summry plot of D0's 1998 angular distribution analysis (R defined in terms of chi)
PNGpng D0_jet_HT_compositeness_PRDRC_62_031101_2000_money_plot.png r1 manage 78.0 K 2010-09-03 - 10:39 AmnonHarel A summry plot of D0's 2000 multijet HT analysis
PNGpng c3lever.png r1 manage 9.1 K 2010-09-03 - 11:02 AmnonHarel  
PNGpng ratio_pythia_MpiHadvsNoMpiHad.png r1 manage 80.4 K 2010-09-03 - 10:21 AmnonHarel  
Edit | Attach | Watch | Print version | History: r4 < r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r4 - 2010-09-05 - AmnonHarel
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback