05 May

May 2009 posts

 

2009.05.03 LDA: varying pt and eta cut

Cut optimization with Fisher's LDA classifier

ROOT implementation for LDA:

Application for cuts optimization in the gamma-jet analysis

LDA configuration: default

LDA input parameters Set0:

  1. Set0:
    • Energy fraction in 3x3 cluster within a r=0.7 radius:
      E_3x3/E_0.7
    • Photon-jet pt balance:
      [pt_gamma-pt_jet]/pt_gamma
    • Number of charge tracks within r=0.7 around gamma candidate:
      Ncharge
    • Number of Endcap towersL fired within r=0.7 around gamma candidate:
      NtowBarrel
    • Number of Barrel towers fired within r=0.7 around gamma candidate
      NtowEndcap
  2. Set1:
  3. Set2:
    • All from Set1
    • Energy fraction in E_2x1 and E_2x2 witin E_3x3:
      E_2x1/E_2x2 and E_2x2/E_3x3
  4. Set3:
    • All from Set2
    • Energy in post-shower layer under 3x3 tower patch:
      E_post^3x3

The number of strips in SMD u or v planes is required to be greater than 3

Pre-shower sorting (energy in tiles under 3x3 tower patch):

  1. pre1=0, pre2=0
  2. pre1=0, pre2>0
  3. 0 < pre1 < 0.004
  4. 0.004 < pre1 < 0.01
  5. pre1 < 0.01
  6. pre1 >= 0.01

Photon pt and rapidity cuts:

  1. pt>7GeV
  2. pt>8GeV
  3. pt>9GeV
  4. pt>10GeV
  5. detector eta <1.4 (pt>7GeV)
  6. detector eta > 1.4 (pt>7GeV)

Figure 0: photon pt distribution for pre-shower1<0.01
Colour coding:
black pp2006 data, red gamma-jet MC, green QCD MC, blue gamma-jet+QCD

LDA Set0

Figure 1: LDA discriminant with Set0: Data to Monte-Carlo comparison (pt>7GeV cut)

Right plot for each pre-shower condition shows the ratio of pp2006 data to sum of the Monte-Carlo samples
Colour coding:
black pp2006 data, red gamma-jet MC, green QCD MC, blue gamma-jet+QCD


Figure 2: efficiency, purity, rejection vs. LDA discriminant (pt>7GeV cut)


Figure 3: rejection vs. efficiency

Figure 4: purity vs. efficiency

Figure 5: purity vs. rejection

LDA Set1

Figure 6: LDA discriminant with Set1: Data to Monte-Carlo comparison


Figure 7: rejection vs. efficiency

Figure 8: purity vs. efficiency

Figure 9: purity vs. rejection (click link to see the figure)

LDA Set2

Figure 10: rejection vs. efficiency (click link to see the figure)

Figure 11: purity vs. efficiency

Figure 12: purity vs. rejection (click link to see the figure)

LDA Set3

Figure 13: rejection vs. efficiency (click link to see the figure)

Figure 14: purity vs. efficiency

Figure 15: purity vs. rejection (click link to see the figure)

2009.05.04 LDA: More SMD info, 3x3 tower energy, correlation matrix

Cut optimization with Fisher's LDA classifier

ROOT implementation for LDA:

Application for cuts optimization in the gamma-jet analysis

LDA configuration: default

LDA input parameters Set0:

  1. Set4 (link for results with LDA Set0-Set3):
    • Energy fraction in 3x3 cluster within a r=0.7 radius:
      E_3x3/E_0.7
    • Photon-jet pt balance:
      [pt_gamma-pt_jet]/pt_gamma
    • Number of charge tracks within r=0.7 around gamma candidate:
      Ncharge
    • Number of Endcap towersL fired within r=0.7 around gamma candidate:
      NtowBarrel
    • Number of Barrel towers fired within r=0.7 around gamma candidate
      NtowEndcap
    • Shower shape analysis: distance to 80% cut line:
      distance to cut line
    • Energy fraction in E_2x1 and E_2x2 witin E_3x3:
      E_2x1/E_2x2 and E_2x2/E_3x3
    • Energy in post-shower layer under 3x3 tower patch:
      E_post^3x3
    • Tower energy in 3x3 patch:
      E_tow^3x3
    • SMD-u energy in 25 central strips:
      E_smd-u^25
    • SMD-v energy in 25 central strips:
      E_smd-v^25
    • SMD-v peak energy (in 5 central strips):
      E_peak

The number of strips in SMD u or v planes is required to be greater than 3

Pre-shower sorting (energy in tiles under 3x3 tower patch):

  1. pre1=0, pre2=0
  2. pre1=0, pre2>0
  3. 0 < pre1 < 0.004
  4. 0.004 < pre1 < 0.01
  5. pre1 < 0.01
  6. pre1 >= 0.01

Integrated yields per pre-shower bin:

sample total integral pre1=0,pre2=0 pre1=0, pre2>0 0 < pre1 < 0.004 0.004 < pre1 < 0.01 pre1 < 0.01 pre1 >= 0.01
photon-jet 2.5640e+03 3.5034e+02 5.2041e+02 5.6741e+02 5.2619e+02 1.9644e+03 5.9994e+02
QCD 5.6345e+04 1.3515e+03 4.3010e+03 1.2289e+04 1.5759e+04 3.3701e+04 2.2644e+04
pp2006 6.2811e+04 6.8000e+02 2.4310e+03 1.2195e+04 1.6766e+04 3.2072e+04 3.0739e+04

Photon pt and rapidity cuts:

  1. pt>7GeV
  2. pt>8GeV
  3. pt>9GeV
  4. pt>10GeV
  5. detector eta <1.4 (pt>7GeV)
  6. detector eta > 1.4 (pt>7GeV)

LDA Set4

Figure 1: LDA discriminant with Set0: Data to Monte-Carlo comparison (pt>7GeV cut)

Right plot for each pre-shower condition shows the ratio of pp2006 data to sum of the Monte-Carlo samples
Colour coding:
black pp2006 data, red gamma-jet MC, green QCD MC, blue gamma-jet+QCD


Figure 2: rejection vs. efficiency

Figure 3: purity vs. efficiency

Figure 4: purity vs. rejection

Figure 5: Correlation matrix (pt>7GeV cut)
pre1=0, pre2=0

pre1=0, pre2>0

0 < pre1 < 0.004

0.004 < pre1 < 0.01

pre1 < 0.01

pre1 >= 0.01

2009.05.06 Applying cuts on LDA: request minimum purity or efficiency

Cut optimization with Fisher's LDA classifier

For this post LDA input parameters Set4 has been used

LDA for various pre-shower bins is trained independetly,
and later results with pre-shower1<0.01 are combined.

There are a set of plots for various photon pt cuts (pt> 7, 8, 9 10 GeV)
and with different selection of cutoff for LDA
(either based on purity or efficiency).
Number in brackets shows the total yield for the sample.

Link to all plots (16 total) as a single pdf file

pt > 7GeV

Figure 1: pt > 7GeV, efficiency@70

Figure 2: pt > 7GeV, purity@35

 

Figure 3: pt > 7GeV, purity@40

 

Figure 4: pt > 7GeV, purity@25 (Note: very similar to results with efficiency@70)

 

pt > 9GeV

Figure 5: pt > 9GeV, efficiency@70

 

Figure 6: pt > 9GeV, purity@35

 

pt > 10GeV

Figure 7: pt > 10GeV, efficiency@70

 

Figure 8: pt > 10GeV, purity@40

 

2009.05.07 Photon-jets analysis with the Endcap Calorimeter

Photon-jets with the Endcap Calorimeter

(analysis status update for Spin PWG)

Slides in pdf format:

 

2009.05.12 Variable distributions after LDA at 70% efficiency

Cut optimization with Fisher's LDA classifier

For this post LDA results with Set1 and Set2 has been used
Note, that LDA for various pre-shower bins is trained independetly

pdf-links with results for pre1=0 and pre2=0 (pre-shower bin 1):

Figures below are for 0.004<pre-shower1<0.01 (pre-shower bin 4).

Photon pt cut: pt> 7, pre-shower bin: 0.004 < pre1 < 0.01
LDA cut with efficiency @ 70%

Set1 vs. Set2

What is added in Set2 compared to Set1:
smaller cluster size information (r2x1, r2x2), post-shower energy

Figure 1: r2x1
before LDA cut

LDA cut for Set1

LDA cut for Set2

Figure 2: r2x2
before LDA cut

LDA cut for Set1

LDA cut for Set2

Figure 3: r3x3
before LDA cut

LDA cut for Set1

LDA cut for Set2

Figure 4: Residual distance
before LDA cut

LDA cut for Set1

LDA cut for Set2

Other variables with LDA Set2 cut

Note: Only plos for LDA cut @70 efficiency for Set2 are shown

Figure : number of charge particles around photon

 

Figure 5: number of EEMC tower around photon

 

Figure 6: number of BEMC tower around photon

 

Figure 7: photon-jet pt balance

 

Figure 8: SMD energy in 5 centrapl strips

 

Figure 9: SMD energy in 25 central strips: u and v plane separately (plot for V plane)

 

Figure 10: 2x1 cluster energy

 

Figure 11: 2x2 cluster energy

 

Figure 12: 3x3 cluster energy

 

Figure 13: tower energy in r=0.7 radius

 

Figure 14: 3x3 pre-shower1 energy

 

Figure 15: 3x3 pre-shower2 energy

 

Figure 16: 3x3 post-shower energy

 

Figure 17: photon pt

 

Figure 18: jet pt

 

Figure 19: z vertex

2009.05.31 CIPANP 2009 photon-jet presentation

CIPANP 2009 presentation on photon-jet study

Title:
"Photon-jet coincidence measurements
in polarized pp collisions at sqrt{s}=200GeV
with the STAR Endcap Calorimeter"