Update on the HZZ 250 GeV +80/-30 250/fb Analysis for the 5 September 2013 SiD Meeting:

Analysis:

  • divide into 4 jet / 6 jet topologies
  • apply preselection depending on topology
  • train/apply TMVA's
  • validate with cut table
  • check remaining backgrounds



Preselection variables:

Evis<140.0: (4 Jet category)
# nTrks>10 &&
# EvisJETS>60.0 &&&nbsp;
# PTvisJETS>5.0 &&&nbsp;
# hmass>50.  && hmass<140. &&&nbsp;
# fabs(fabs(dph)-3.14159)>0.15  &&&nbsp;
# ej1>40. &&&nbsp;
# jetthrust<0.99

Evis>140: (6 jet category)

# nTrks>20 &&
# EvisJETS<220.0 &&
# PTvisJETS>5.0  &&
# hmass>100. && hmass<150. &&
# fabs(fabs(dph)-3.14159)>0.05 &&
# ej1<130. &&
# jetthrust<0.99

The TMVA variables:

  1. Evis JETS
  2. Mvis JETS
  3. PTvis JETS
  4. B Likeness
  5. C likeness
  6. Number of HE Electrons
  7. Higgs Mass
  8. Z energy
  9. Zstr energy
  10. Z cos theta
  11. Zstr cos theta
  12. Z mass
  13. Zstr mass
  14. azimuthal separation of Z jets
  15. thrust
  16. Number Charged Tracks
  17. Number of electrons
  18. Number of muons

    Performance of different MVA options:

 [neal@localhost weights]$ grep -A 8 -i optimal- ../hzzv14-presel-v4-new-common0410results.txt
 --- Classifier   (  #signal, #backgr.)  Optimal-cut  S/sqrt(S+B)      NSig      NBkg   EffSig   EffBkg
 -- ---------------------------------------------------------------------------------------------
 ---       Cuts:  ( 664.0744,1312202.5)      -0.0050            0         0         0        0        0
 --- Likelihood:  ( 664.0744,1312202.5)       1.0000     0.728991  644.0848  779980.1   0.9699   0.5944
 ---     Fisher:  ( 664.0744,1312202.5)       0.0120      1.70335  258.2099  22721.13   0.3888  0.01732
 ---       BDTG:  ( 664.0744,1312202.5)      -0.9823      1.17754  538.5353  208621.3    0.811    0.159
 ---        BDT:  ( 664.0744,1312202.5)      -0.0462      3.42709  100.6751  762.2891   0.1516 0.0005809
 -- ---------------------------------------------------------------------------------------------

Cut table for BDT:

isigall= 1345.984602 isigpresel= 1460.101057 igood= 114.116455 ibad= 762.500000
intot= 77692005.984714 ipre= 876.616455
cuts__ (Cut Name            ):       all   others       2f       4f       6f       aa       1f       3f       5f        q     e+mu      tau       nu  Signif.
cut #0 (all                 ): 7.769e+07   625000 14507062  1019938        0 54073784        0  7464875        0      948       91       45      262      0.15  Delta(sig*BR) = 6.548594 +/- 0.046879
cut #1 (20.<PTvisJETS       ): 1.283e+07    37500  5525512   746188        0  2968250        0  3546875        0      806       69       44      259      0.33  Delta(sig*BR) = 3.042092 +/- 0.024172
cut #2 (EvisJETS<220.       ): 4.637e+06    25000  3287550   514400        0   130196        0   678875        0      608       42       41      219      0.42  Delta(sig*BR) = 2.365748 +/- 0.020248
cut #3 (50.<hmass<140.      ): 3.306e+06        0  2542562   378675        0    60426        0   323875        0      536       38       35      206      0.45  Delta(sig*BR) = 2.233245 +/- 0.019503
cut #4 (||dph|-3.14159|>0.15 ): 2.772e+06        0  2083912   364150        0    57097        0   266000        0      514       36       33      202      0.47  Delta(sig*BR) = 2.120439 +/- 0.018965
cut #5 (nTrks>10            ): 1.426e+06        0  1136375   169088        0    30283        0    89875        0      466       10       19      177      0.56  Delta(sig*BR) = 1.777889 +/- 0.017994
cut #6 (jetthrust<0.99      ): 1.325e+06        0  1044700   165588        0    30033        0    84125        0      463       10       19      173      0.58  Delta(sig*BR) = 1.730704 +/- 0.017728
cut #7 (ej1>40.             ): 1.313e+06        0  1035112   164625        0    29215        0    83250        0      462       10       19      173      0.58  Delta(sig*BR) = 1.725415 +/- 0.017704
cut #8 (MVA                 ): 8.766e+02        0      138      625        0        0        0        0        0       51        0        1       62      3.85  Delta(sig*BR) = 0.259452 +/- 0.015456

Remaining backgrounds:

$ sed 's/\./,/g' zzhpassing.dat | awk '{if (NF==18) print $3,$7,$15}' | sort -n | uniq -c | awk '{if (NF==4) aa+=($1*$4);print $1*$4"\t"$0} END{print "Sum of weights = "aa}' | sort -k 1,1 -n
Sum of weights = 762,5
12,5          1 106562 4f_sz 12,500000
12,5          1 106566 4f_sw 12,500000
12,5          1 106572 4f_sz 12,500000
12,5          1 106575 4f_zz 12,500000
12,5          1 106607 2f_z_ 12,500000
25            2 106551 4f_ww 12,500000
25            2 106553 4f_zz 12,500000
25            2 106571 4f_sz 12,500000
25            2 106573 4f_zz 12,500000
37,5          3 106574 4f_zz 12,500000
37,5          3 106578 4f_ww 12,500000
125          10 106608 2f_z_ 12,500000
162,5        13 106576 4f_zz 12,500000
237,5        19 106577 4f_ww 12,500000

Background contributions for Evis>140:

[neal@localhost weights]$ sed 's/\./,/g' zzhpassing.dat | awk '{if ($18>=140. && NF==18) print $3,$7,$15}' | sort -n | uniq -c | awk '{if (NF==4) aa+=($1*$4);print $1*$4"\t"$0} END{print "Sum of weights = "aa}' | sort -k 1,1 -n
Sum of weights = 187,5
12,5          1 106551 4f_ww 12,500000
12,5          1 106577 4f_ww 12,500000
25            2 106553 4f_zz 12,500000
25            2 106573 4f_zz 12,500000
25            2 106576 4f_zz 12,500000
37,5          3 106574 4f_zz 12,500000
50            4 106608 2f_z_ 12,500000

Background contributions for Evis<140:

 [neal@localhost weights]$ sed 's/\./,/g' zzhpassing.dat | awk '{if ($18<140. && NF==18) print $3,$7,$15}' | sort -n | uniq -c | awk '{if (NF==4) aa+=($1*$4);print $1*$4"\t"$0} END{print "Sum of weights = "aa}' | sort -k 1,1 -n
Sum of weights = 575
12,5          1 106551 4f_ww 12,500000
12,5          1 106562 4f_sz 12,500000
12,5          1 106566 4f_sw 12,500000
12,5          1 106572 4f_sz 12,500000
12,5          1 106575 4f_zz 12,500000
12,5          1 106607 2f_z_ 12,500000
25            2 106571 4f_sz 12,500000
37,5          3 106578 4f_ww 12,500000
75            6 106608 2f_z_ 12,500000
137,5        11 106576 4f_zz 12,500000
225          18 106577 4f_ww 12,500000 <***************

 --- Classifier   (  #signal, #backgr.)  Optimal-cut  S/sqrt(S+B)      NSig      NBkg   EffSig   EffBkg
--- --------------------------------------------------------------------------------------------------
---       Cuts:  (309.89453,  1128865)      -0.0050            0         0         0        0        0
--- Likelihood:  (309.89453,  1128865)       1.0000     0.378385  299.2573  625191.7   0.9657   0.5538
---     Fisher:  (309.89453,  1128865)       0.0088      1.63217  150.0501  8301.587   0.4842 0.007354
---       BDTG:  (309.89453,  1128865)      -0.9950      1.40141  134.4158  9065.234   0.4337  0.00803
---        BDT:  (309.89453,  1128865)       0.0177      3.63521   38.5486  73.90136   0.1244 6.547e-05
--- --------------------------------------------------------------------------------------------------

after raising reco higgs mass cut from 50 GeV to 100 GeV:

 --- Classifier   (  #signal, #backgr.)  Optimal-cut  S/sqrt(S+B)      NSig      NBkg   EffSig   EffBkg
--- --------------------------------------------------------------------------------------------------
---       Cuts:  (209.30833,    51800)      -0.0050            0         0         0        0        0
--- Likelihood:  (209.30833,    51800)       0.9105      1.08931  191.9554  30860.64   0.9171   0.5958
---     Fisher:  (209.30833,    51800)       0.0176      1.78558  116.8125  4162.945   0.5581  0.08037
---       BDTG:  (209.30833,    51800)      -0.5098      1.00598  204.0195  40926.24   0.9747   0.7901
---        BDT:  (209.30833,    51800)       0.0361      2.92564   72.9263  548.4119   0.3484  0.01059
--- --------------------------------------------------------------------------------------------------

Evis<140 GeV and only idrup = 106577 for the background:



Potential tools for reducing the 4f_WW background:

  • higher cut on number of charged tracks (unlikely to help S)
  • extra cuts but keep any event that has high B-likeness
  • redoing n-tuples adding variables for jet combos consistent with W decays (IN PROGRESS)

Note concerning ZZ background: wider cut on acoplanarity should help

  • No labels