Документ взят из кэша поисковой машины. Адрес оригинального документа : http://zebu.uoregon.edu/~uochep/talks/talks04/Lu12-04.pdf Дата изменения: Tue Dec 21 19:59:04 2004 Дата индексирования: Tue Oct 2 10:37:04 2012 Кодировка: Поисковые слова: m 81

Skim Study in CM2
Minghui Lu
University of Oregon

0

For Fully Inclusive

b s group

Notre Dame-Oregon-Pisa-SLAC-UCSC

0 Skim Study in CM2 ­ p.1


Outline
Introduction
0 skim cuts 0 mass and energy spectrum

The Comparison of 0 mass spectrum between SP5 MC and run1 data Preliminary 0 mass fit Outlook

0 Skim Study in CM2 ­ p.2


Introduction
0 is the dominant background for b->s signal in Ї generic B B events 0 correction receipe was provided by Neutral group based on events for SP5 and SP6 (BAD870), but similar study is not done with inclusive events

This is a first try to migrate current inclusive b->s analysis to CM2 To understand 0 spectrum in CM2, provide useful information for other CM2 analyses

0 Skim Study in CM2 ­ p.3


skim
Skim Cuts BGFMultiHadron tag bit 1.0 < e1Mag <3.5 GeV 2nd Fox-Wolfram moment in overall CMS, R2<0.9 nTrk>2, from nGoodTrackLoose p1Mag<2.5 GeV Skim Efficiency Ї B 0 B 0 MC events: 200k ->25,393, 12.7% B + B - MC events: 200k -> 26,417, 13.2% On-Peak run1 data: 500k ->45,395, 9.1% Off-Peak run1 data: 100k -> 7,490, 7.5%

0

0 Skim Study in CM2 ­ p.4


Selection of Candidate
2 Mass cut:
E
1

0

> 1.0 GeV, from user defined HE List

E2 > 30 MeV, from GoodPhotonLoose List 50 < m2 < 250 MeV 1.0 < E2 < 3.5 GeV

Truth matching for MC: GHit truth map The mother <->daughter matching

0 Skim Study in CM2 ­ p.5


spectrum (on-peak data)
Mpi0
Entries Mean RMS Underflow Overflow Integral

0

h1d3
95083 0.1567 0.04916 0 0 9.508e+04

Epi0
Entries Mean RMS

h1d4
95083 1.59 0.4771 0 0 9.508e+04

6000 5000 4000 3000 2000 1000 0

6000 5000 4000 3000 2000 1000 0 1

Underflow Overflow Integral



0

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

1.5

2

2.5

3

3.5

spectrum will be studied in 8 different E 1.0-1.2, 1.2-1.4, 1.4-1.6, 1.6-1.8, 1.8-2.0, 2.0-2.2, 2.2-2.4, 2.4-3.0



0

bins:

0 Skim Study in CM2 ­ p.6


M0 in differnt E bins(on-peak data)
0

0 Skim Study in CM2 ­ p.7


Ї spectrum for B B 0 MC
Pi0Mass
Entries 56111 Mean 0.1607 RMS 0.05082

0

0

Pi0cmE

Entries 56111 Mean RMS 1.395 0.3076

Mpi0 {Mpi0>0.05 && Mpi0<0.25}

Entries 14254 Mean 0.1323

2000

3500

1800

RMS 0.01784

1800

1600 3000 1400 2500

1600

1400

1200

1200

2000

1000

1000 1500 800 600 600 1000 400 400 500 200 200 800

0 0.05 0.1 0.15 0.2 0.25

0 1 1.5 2 2.5 3 3.5

0 0.05 0.1 0.15 0.2 0.25 0 Skim Study in CM2 ­ p.8


Ї M0 in 8 different E bins (B B 0)
0

0

0 Skim Study in CM2 ­ p.9


MC and Run1 data comparison
on-off: On-peak Lon L0f f

off-peak

Ї Combine both B 0 B 0 and B + B - , normalized to same number of Bs

Events/4MeV

7000 6000 5000 4000 3000 2000 1000 0

0 spectrum
Entries Mean RMS 112222 0.1607 0.05082

0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24 2

0 Skim Study in CM2 ­ p.10


mass fit function
The signal Model:
f (m) = Ag [f1 G(m, µ1 , 1 ) + (1 - f1 )G(m, µ2 , 2 )], m > m f (m) = N
p1 / (m0 -m)+p1 / 1 p 0

0

,m < m

0

where m0 = µ1 - Background:
f (m) =
amb (m2 +c)
d

8 fit parameters: Ag , f1 , µ1 ,1 , µ2 , 2 , p and .

4 fit parameters: a, b, c, d

0 Skim Study in CM2 ­ p.11


Preliminary mass fit
Mpi0Truth
114.5 / 42 2 / ndf fraction 0.9109 ± 0.0066 area 43.76 ± 0.71 mean1 0.1347 ± 0.0001 sigma1 0.006113 ± 0.000086 mean2 -0.07278 ± 0.02710 15.61 ± 0.45 sig2/sig1 lambda 1.246 ± 0.043 1.915 ± 0.115 poly

0

103

102

10

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0 Skim Study in CM2 ­ p.12


Summary and Outlook
A simple data MC comparison was done on pi0 skim Preliminary Mass fits on truth matched 0 mass spectrum were performed, more complete mass fits are expected Only skim samples were studied so far, more cuts will be implemented, such as photon quality cut, fisher cut, etc The mass fits now are done by my personal codes, John and Francesca's complete 0 / mass fit package will be borrowed and migrated to CM2 This study is intended to be done on full statistics (run1 to run4 data) once all cuts are finalized and codes are implemented
0 Skim Study in CM2 ­ p.13