Revisiting the Top Background (and a little of B-tagging) G. - PowerPoint PPT Presentation

May’12 Higgs 2l2q Working Meeting Revisiting the Top Background (and a little of B-tagging) G. Codispoti, J.F. de Trocóniz Universidad Autónoma de Madrid

Outline q Using eµ data for top(+X) background: full 2011 results. q First checks of 2012 eµ data: comparing first 674 pb -1 to 2011 sample. q Observations / doubts about B-tagging with JP.

Description of the 2011 eµ + jj Sample M(eµ) > 50 GeV eµ in Z mass window q Sample composition: White: 86% TTbar (X-section +4%) q Blue: 6% fakes (reversing lepton isolation cut); 8% DY (taus) + single top + WW

Description of the 2011 eµ + jj eµ in Z mass window eµ and jj in Z mass window q Sample composition: 86% TTbar; 6% fakes; 8% DY (taus) + single top + WW

Description of the 2011 eµ + jj ≥ 1 TCHEM tag 2 (TCHEM + TCHEL) tags MET Significance < 10. q Sample composition: 89(94)% TTbar; 6(3)% fakes; 5(3)% other small backgr.

Top(+X) Background in DY + jj Sample q eµ and ee/µµ top data are symmetric in normalization and shape: predicted by MC and confirmed using high-MET and/or out-of-Z- mass-window data. 2011 ee/µµ data 2011 ee/µµ data ≥ 1 TCHEM tag ≥ 1 TCHEM tag 2011 eµ data 2011 eµ data

Top(+X) Background in DY + jj Sample q eµ and ee/µµ top data are symmetric in normalization and shape: predicted by MC and confirmed using high-MET and/or out-of-Z- mass-window data. 2011 ee/µµ data 2011 ee/µµ data 2 (TCHEM + TCHEL) tags 2 (TCHEM + TCHEL) tags 2011 eµ data 2011 eµ data

Top(+X) Background in DY + jj Sample 2011 ee/µµ data 2011 ee/µµ data ≥ 1 TCHEM tag ≥ 1 TCHEM tag leptonic Z sidebands leptonic Z sidebands 2011 eµ data 2011 eµ data q Is DATA; is available; is understood. q We might consider using a top(+X) data-driven background subtraction in the signal and sideband mass regions; the alpha correction would be restricted to the DY background.

eµ + jj Data: 2011 vs. 2012

eµ + jj Data: 2011 vs. 2012

eµ + jj Data: 2011 vs. 2012

eµ + jj Data: 2011 vs. 2012 ≥ 1TCHEM tag 2 (TCHEM + TCHEL) tags

SF for TC HE: Light Jets q Light flavor enriched jet sample (2011 ee and µµ data). q Jets ordered by TCHE discriminator value. < 0 0 - L L - M M < 0 0 - L L-M M TCHE Discr. For 1 st Jet TCHE Discr. For 2 nd Jet q Standard efficiencies and SFs.

SF for TC HE: B-jets q B-enriched jet sample (2011 eµ data). q Jets ordered by TCHE discriminator value. < 0 0 - L L - M M < 0 0 - L L-M M TCHE Discr. For 1 st Jet TCHE Discr. For 2 nd Jet q Standard efficiencies and SFs.

SF for Jet Probability: Light Jets q Light flavor enriched jet sample (2011 ee and µµ data). q Jets ordered by JP discriminator value. < 0 0 - L L - M M < 0 0 - L L-M M JP Discr. For 1 st Jet JP Discr. For 2 nd Jet q Standard efficiencies and SFs.

SF for Jet Probability: B-jets q B-enriched jet sample (2011 eµ data). q Jets ordered by JP discriminator value. < 0 0 - L L - M M < 0 0 - L L-M M JP Discr. For 1 st Jet JP Discr. For 2 nd Jet q Standard efficiencies and SFs.

SF for Jet Probability: B-jets q B-enriched jet sample (2011 eµ data). q Jets ordered by JP discriminator value. < 0 0 - L L - M M < 0 0 - L L-M M JP Discr. For 1 st Jet JP Discr. For 2 nd Jet q For B-jets it’s much better to use SF_b = 1. (?!)

BACKUP

Beta Distribution for Taggeable Jets q 2011 ee and µµ data. q Jet taggeability condition as presented by Giuseppe in previous meeting (valid for both TC and JP).