news from cteq-tea pdf analysis · ¥ nnlo cteq global analysis (in progress) i validation of o(α2...

News from CTEQ-TEA PDF analysis

Pavel Nadolsky

Southern Methodist UniversityDallas, TX, U.S.A.

in collaboration withM. Guzzi, J. Huston, H.-L. Lai, Z. Li,

J. Pumplin, D. Stump, and C.-P. Yuan

April 11, 2010

Pavel Nadolsky (SMU) XIX DIS workshop, Newport News April 11, 2010 1

CTEQ-Tung Et Al.: ongoing activities

¥ CT10 and CT10W general-purpose NLO PDF sets

¥ NNLO CTEQ global analysis (in progress)

I Validation of O(α2s) heavy-quark contributions to DIS is

completed (details by M. Guzzi in the HQ WG session)

¥ Effects of new experimental data on PDFsI W lepton asymmetry, Fn

2 /F p2 , comparisons with the LHC data

I dependence on the PDF parametrization form (J. Pumplin,today)

¥ DGLAP factorization in DIS at small x


CT10 parton distribution functions (PRD82, 074024 (2010))

¥ General-purpose NLO PDFs

¥ Adequate for the majority ofPQCD applications

¥ includes combined HERA-1 DISand Tevatron Run-2 inclusive jetdata

¥ detailed analysis of the TevatronRun-2 W asymmetry (A`) data

I CT10 and CT10W sets, withdifferent treatment of A`

¥ no LHC data yet10-3

10-2

10-1

1

10

102

103

104

105

106

107

10 102 103 104 105

σNC

r (

x,Q

2 ) ⋅

2i e+ P

→ e

+ X

Q2 [GeV2 ]

x=6.18⋅ 10-5 , i=20

x=9.5⋅ 10-5 , i=19x=2⋅ 10-4 , i=18

x=3.2⋅ 10-4 , i=17x=5⋅ 10-4 , i=16

x=8⋅ 10-4 , i=15

x=1.3⋅ 10-3 , i=14

x=2⋅ 10-3 , i=13

x=3.2⋅ 10-3 , i=12

x=5⋅ 10-3 , i=11

x=8⋅ 10-3 , i=10

x=1.3⋅ 10-2 , i=9

x=2⋅ 10-2 , i=8

x=3.2⋅ 10-2 , i=7

x=5⋅ 10-2 , i=6

x=8⋅ 10-2 , i=5

x=0.13, i=4

x=0.18, i=3

x=0.25, i=2

x=0.4, i=1

x=0.65, i=0

Shifted HERA-1 data (circles)CT10 NLO theory (lines)


CT10 parton distribution functions (PRD82, 074024 (2010))

¥ 53 CT10/CT10W eigenvector sets for αs(MZ) = 0.118

I 4 CT10AS/CT10WAS PDFs for αs(MZ) = 0.116− 0.120

♦ The correlated PDF+αs uncertainty on an observable X iscomputed by

∆XPDF+αs =q

∆X2PDF, CT10 + ∆X2

αs, CT 10AS ,

as explained in PRD 82,054021 (2010)

I CT10/CT10W PDFs with 3 and 4 active flavors

¥ In the LHADPF library and at http://hep.pa.msu.edu/cteq/public/index.html


NNLO global PDF analysis¥ Time is ripe for producingNNLO PDFs

¥ Accuracy of many EW, DIS, jetdata becomes comparable toNNLO contributions

¥ Heavy-quark mass effects inDIS at Q ≈ mQ is the key chal-lenge for NNLO global fits

⇒ 5-7% differences in σW,Z

at the LHC (Tung et al., hep-ph/0611254)

¥ For comparison, NNLO hard-scattering correction to σW,Z is≈2%

18.5 19 19.5 20 20.5 21 21.5 22ΣtotHpp®HW±®{ΝLXL HnbL

1.85

1.9

1.95

2

2.05

2.1

2.15

Σto

tHpp®HZ

0®{{-

LXLHn

bL

W± & Z cross sections at the LHC

CTEQ6.6 (GM)

CTEQ6.1 (ZM)

NNLL-NLO ResBos

G. Watt, PDF4LHC mtg, 26.03.2010


NNLO global PDF analysis

NNLO DIS contributions are fullyimplemented in the S-ACOTscheme, the defaultfactorization scheme ofCTEQ6.6 and CT10 PDFs

18.5 19 19.5 20 20.5 21 21.5 22ΣtotHpp®HW±®{ΝLXL HnbL

1.85

1.9

1.95

2

2.05

2.1

2.15

Σto

tHpp®HZ

0®{{-

LXLHn

bL

W± & Z cross sections at the LHC

CTEQ6.6 (GM)

CTEQ6.1 (ZM)

NNLL-NLO ResBos



Simplified Aivazis-Collins-Olness-Tung schemeACOT, PRD 50 3102 (1994); Collins, PRD 58 (1998) 094002; Kramer, Olness, Soper, PRD (2000) 096007

¥ Derivation is based upon, and closely follows, the proof ofQCD factorization for DIS with massive quarks (Collins, 1998)

¥ Relatively simple

I One value of Nf (and one PDF set) in each Q range

I Straightforward matching based on kinematical rescaling

I Sets mQ = 0 in ME with incoming c or b

¥ Reduces to the ZM MS scheme at Q2 À m2Q, without

additional renormalization

¥ Reduces to the FFN scheme at Q2 ≈ m2Q

I has reduced dependence on tunable parameters at NNLO


Input parameters of the S-ACOT schemeAt NLO, the charm mass mc, factorization scale µ, and rescalingvariable ζ of CTEQ PDFs are tuned to best describe the DIS data

ç

ç

çç

ç

à

à

à

à

à

æ

æ

æ æ

æ

ò

ò

òò

ò

S-ACOT band, from top:

Μ2=Q2+4mc

2,Λ=0.2Μ

2=Q2+mc

2,Λ=0 (default)Μ

2=Q2,Λ=0

Dashed: NF=3, NNLO

ç MSTW08-NLOà MSTW08-NLO-Χò FONNL-A-Χæ FONLL-B-Χ

10-5 10-4 10-3 0.01 0.02 0.05 0.1 0.20.0

0.5

1.0

1.5

2.0

2.5

3.0

x

103x0.

5F

2c H

x,QL

NLO, Q = 2 GeV, mc = 1.41 GeV

2009 Les Houches HQ benchmarkswith toy PDFs; default µ = Q


W, Z cross sections;mc = 1.3 GeV in CTEQ6.6


NNLO results for F(c)2 (x,Q2) - Preliminary

At NNLO and Q ≈ mc:

S-ACOT-χ (Nf = 4) ≈ FFN (Nf = 3)

without tuning

¥ S-ACOT is numerically closeto other NNLO schemes

¥ NNLO expressions are closeto the FONLL-C scheme(Forte, Laenen, Nason, arXiv:1001.2312).

ò

ò

ò

ò

æ

æ

æ æ

æ

scale dependence

blue: S-ACOT-Χ NLO

green: S-ACOT-Χ NNLO

magenta: FFNS NNLO Nf=3

ò MSTW08-NNLO-Χ

æ FONNL-C-Χ

10-5 10-4 10-3 0.01 0.02 0.05 0.1 0.20.0

0.5

1.0

1.5

2.0

2.5

x

103x0.

5F

2c H

x,QL

LH PDFs Q=2 GeV, mc=1.414 GeV

¥ ACOT formalism provides recipe-like formulasfor implementing NNLO in the GM scheme (⇒Guzzi)


New: F(c)2 (x,Q2) in S-ACOT scheme at NNLO

Preliminary

GM ACOT-Χ NNLO

FFNS Nf=3 NNLO

ZM NNLO

0.00

0.05

0.10

0.15

0.20

0.25

F2

hHx

,QL

x=0.01

10.5.2. 20.3. 30.1.5 15.7.0.60.70.80.91.01.11.21.3

Q HGeVL

Rat

ioto

GM

NNLO calculationfor F c

2,L(x,Q) isimplemented inthe CTEQ fit (Guzzi, Lai,

P.N., Yuan, in preparation)

ACOT reducesto FFNS at Q ≈ mc

and to ZM at Q À mc

Les Houches toyPDFs, evolved at

NNLO withthreshold matching

terms


CTEQ PDFs vs. the latest data: LHCAgreement with many LHC measurements

Figures are from ATLAS. Similar results from CMS

+data on σW /σZ , tt, γγ, etc.


CTEQ PDFs vs. the latest data: Tevatron

Outstanding puzzles:1. Run-2 W charge asymmetry

(constraining d(x,Q)/u(x,Q)at x > 0.1)

2. Inclusive (di)jet production(constraining g(x,Q) atx > 0.1)

In both processes, experimentalaccuracy is high enough to startfeeling effects beyond NLO andof resummations

ey

0 0.5 1 1.5 2 2.5 3

)e

(y

eA

-0.8

-0.6

-0.4

-0.2

0

0.2

> 25 GeVeT

p

=1.96 TeVS+X ν ± e→ ± W→pp

> 25 GeVνTE

)-1D0 electron data (0.75 fbCT10W (Solid band)CTEQ6.6 (Hatched band)

(GeV)JJM200 400 600 800 1000 1200 1400

JJ

/dM

σR

ati

o o

f d

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

< 0.4max

|y|

Scale = 0.5 average pT of jets

CTEQ6.6 / CT10.00

CT10 / CT10.00

CT10W / CT10.00


The puzzle of the CDF/D0 W lepton asymmetry

¥ CT10W set reasonably agrees with 3 pT` bins of Ae(ye) andone bin of Aµ(yµ) from D0 Run-2 (2008).

¥ NNPDF 2.0 (arXiv:1012.0836) agrees with Aµ(yµ), disagrees with twopTe bins of Ae(ye).

¥ CT10, many other PDFs fail.

Agreement of Source orPQCD with D0 Ae(ye) χ2/npt comments

CTEQ6.6, NLO 191/36=5.5 Our study;

CT10W, NLO 78/36=2.2 Resbos, NNLL-NLO

With Aµ(yµ): 88/47=1.9ABKM’09, NNLO 540/24=22.5 Catani, Ferrera, Grazzini,

MSTW’08, NNLO 205/24=8.6 JHEP 05, 006 (2010)

JR09VF, NNLO 113/24=4.7


Why difficulties with fitting A`(y`)?

1. A`(y`) is very sensitive to the average slope sdu ofd(x, MW )/u(x, MW )

A`(y`) ∼ A`(yW )|LO ∝1

x1 − x2

[d(x1)

u(x1)− d(x2)

u(x2)

]; x1,2 =

Q√se±yW

Berger, Halzen, Kim, Willenbrock, PRD 40, 83 (1989); Martin, Stirling, Roberts, MPLA 4, 1135 (1989); PRD D50, 6734 (1994);Lai et al., PRD 51, 4763 (1995)

2. Constraints on sdu by fixed-target F d2 (x,Q)/F p

2 (x,Q) areaffected by nuclear and higher-twist effectsAccardi, Christy, Keppel, Monaghan, Melnitchouk, Morfin, Owens, PRD 81, 034016 (2010)


Challenges with fitting A`(y`)

Small changes in sdu causesignificant variations in A`

Lai et al., PRD 51, 4763 (1995)

Alternative constraints on d/uby F d

2 (x,Q)/F p2 (x,Q) from

fixed-target DIS are affectedby nuclear and higher-twisteffectsAccardi, Christy, Keppel, Monaghan, Melnitchouk,Morfin, Owens, PRD 81, 034016 (2010)


d(x,Q)/u(x,Q) at Q = 85 GeV

Solid band: CTEQ6.6 uncertaintyHatched band: CT10 uncertainty

10-5 10-4 10-3 0.01 0.02 0.05 0.1 0.2 0.5 0.70.7

0.8

0.9

1.0

1.1

1.2

1.3

x

Rat

ioto

CT

EQ

6.6M

d/u at µ = 85 GeV

Solid band: CTEQ6.6 uncertaintyHatched band: CT10W uncertainty

10-5 10-4 10-3 0.01 0.02 0.05 0.1 0.2 0.5 0.70.7

0.8

0.9

1.0

1.1

1.2

1.3

x

Rat

ioto

CT

EQ

6.6M

d/u at µ = 85 GeV

¥ CT10W prefers a larger slope of d/u, has a smaller uncertaintythan CTEQ6.6 or CT10

¥ CT10W shows tension with NMC, BCDMS F p,d2 data



3. Existing parametrizations underestimate the PDF uncertaintyon d/u

PDFs based on Chebyshovpolynomials improveagreement with D0 Run-2 Ae,but are outside of currentCTEQ/MSTW bands (Pumplin)

This ambiguity is reduced byA`(y`) at the LHC, whichconstrains d/u and d/u atx ∼ 0.01.

Band: CT10 uncertainty

Long dash: Chebyshov, dbar/ubar->1 at x->0

Short dash: Chebyshov, free dbar/ubar at x->0

Solid: MSTW’2008NLO

PRELIMINARY



4. Experimental A` with lepton pT` cuts is sensitive to dσ/dqT of Wboson at transverse momentum qT → 0.

¥ Fixed-order (N)NLO calculations (DYNNLO, FEWZ, MCFM,...)predict a wrong shape of dσ/dqT at qT → 0.

¥ Small-qT resummation correctly predicts dσ/dqT in this limit.

¥ CT10(W) PDFs are fitted using a NNLL-NLO+K resummedprediction for A` (ResBos); must not be used with fixed-orderpredictions for A`.

For example:

χ2(CT10W+ResBos) = 1.9Npt (us);

χ2(CT10W+DYNNLO) = 8.4Npt (NNPDF)


Charge asymmetry in peT bins (CDF Run-2, 207 pb−1)

Without the peT cut (FEWZ):

Anastasiou et al., 2003

y

y of W boson

With pTe cuts imposed, Ach(ye) is sensitive to small-QT resummation

0.5 1 1.5 2 2.5 3y{

-0.6

-0.4

-0.2

0

0.2

Cha

rge

asym

met

ry

25 < pT{ < 35 GeV, CTEQ65

Resummed

NLOLO

0.5 1 1.5 2 2.5 3y{

-0.4

-0.2

0

0.2

0.4

Cha

rge

asym

met

ry

35 < pT{ < 45 GeV, CTEQ65

NNLL�NLO

NLO

LO

PN, 2007, unpublished; arXiv:1101.0561


CT10(W) vs. A` at the LHC

|µη|0 0.5 1 1.5 2

µA

0.15

0.2

0.25

0.3

0.35

|µη|0 0.5 1 1.5 2

µA

0.15

0.2

0.25

0.3

0.35=7 TeV)sData 2010 (

MC@NLO, CTEQ 6.6MC@NLO, HERA 1.0MC@NLO, MSTW 2008

-1 L dt = 31 pb∫

νµ →W

ATLAS

CT10(W) agrees well with theLHC A`; some differencesbetween NLO and NNLL+NLO

|η|0 0.5 1 1.5 2 2.5 3

W c

har

ge

asym

met

ry

0.1

0.15

0.2

0.25

0.3

0.35 CT10

CT10W

CMS electron

CMS muon

NNLL−NLO+K, ResBos

PRELIMINARY

Zhao Li, 2011


CT10(W) vs. A`: LHC-B

|η|1.5 2 2.5 3 3.5 4 4.5

W c

harg

e as

ymm

etry

-0.6

-0.4

-0.2

0

0.2

0.4

CT10

CT10W

LHCBNNLL−NLO+K, ResBos

PRELIMINARY

Zhao Li, 2011

LHC-B marginally prefers CT10W


Acut fits to combined HERA data

x-510 -410 -310 -210 -110 1

]2 [

GeV

2 T /

p2

/ M

2Q

1

10

210

310

410

510

610 NMC-pdNMCSLACBCDMSHERAI-AVCHORUSFLH108NTVDMNZEUS-H2DYE886CDFWASYCDFZRAPD0ZRAPCDFR2KT

= 0.5cutA = 1.0cutA = 1.5cutA = 3.0cutA = 6.0cutA

NNPDF2.0 dataset

Fitting procedure:¥ Include only DIS data above an Acut line

¥ Compare the resulting PDFs with DIS data below the Acut

line, in a region that is “connected” by DGLAP evolutionPavel Nadolsky (SMU) XIX DIS workshop, Newport News April 11, 2010 20

CT10: Acut fits to DIS data at Q > Q0 = 2 GeV>2

<d

0.5

1

1.5

2

2.5

3 = 1.5cutFit with A

Fit without cuts < 1.0 cutA

< 1.5cut1.0 < A

< 3.0cut1.5 < A

< 6.0 cut3.0 < A

> 6.0cutA

Caola, Forte, Rojo, arXiv:1007.5405v2

Q > 1.41 GeV

<1.0 1.0-1.5 1.5-3.0 3.0-6.0 >6.0

ææ

æ

ææ

à

à

à

àà

ì

ì

ì

ìì

CT10 fit

Fit 1 with Acut=1.5Fit 2 with Acut=1.5

Q > 2 GeV

N pts= 36 18 30 27 468Ags

ranges

0.5

1.0

1.5

2.0

2.5

3.0Χdata

2�Npts

Motivation

Search for deviations from DGLAP evolution at smallest x and Q

¥ Follow the procedure proposed by NNPDF (Caola, Forte, Rojo, arXiv:1007.5405)



<d

0.5

1

1.5

2

2.5



< 1.5cut1.0 < A

< 3.0cut1.5 < A

< 6.0 cut3.0 < A

> 6.0cutA


Q > 1.41 GeV

<1.0 1.0-1.5 1.5-3.0 3.0-6.0 >6.0

ææ

æ

ææ

à

à

à

àà

ì

ì

ì

ìì

CT10 fit


Q > 2 GeV

N pts= 36 18 30 27 468Ags

ranges

0.5

1.0

1.5

2.0

2.5

3.0Χdata

2�Npts

CT10

Two CT10-like fits to data at Ags > 1.5, with differentparametrizations of g(x,Q)

χ2i =

(Shifted Data− Theory)2

σ2uncor

Large syst. shifts at Ags < 1.0, in apattern that could mimic aslower Q2 evolution



<d

0.5

1

1.5

2

2.5



< 1.5cut1.0 < A

< 3.0cut1.5 < A

< 6.0 cut3.0 < A

> 6.0cutA


Q > 1.41 GeV

<1.0 1.0-1.5 1.5-3.0 3.0-6.0 >6.0

ææ

æ

ææ

à

à

à

àà

ì

ì

ì

ìì

CT10 fit


Q > 2 GeV

N pts= 36 18 30 27 468Ags

ranges

0.5

1.0

1.5

2.0

2.5

3.0Χdata

2�Npts

CT10, cont.

δχ2 ∼ 0 at Ags > 1.0[no difference]

δχ2 = 0− 1.5 at Ags < 1.0,with large uncertainty

⇒ Disagreement with the “DGLAP-connected” data atAgs < Acut is not supported by the CT10 fit


Conclusions

¥ In the CTEQ-TEA fit, an NNLO calculation for F c,b2,L in the

S-ACOT scheme is demonstrated to be viable.

¥ This is the most challenging component of the NNLO CTEQPDF analysis, to be made available soon.

¥ Progress in understanding of new Tevatron and LHC datasets, PDF parametrization issues

¥ arXiv:1101.0561: synopsis of recent CTEQ-TEA publications

I CT10W fit to Run-2 W charge asymmetry;PDFs for leading-order showering programs; constraints oncolor-octet fermions


CT10 & CT10W predictions for the near futureTotal cross sections

0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2

+H

-HW

+HW

HZ

t (s-channel)

-W

+W

’ (300)+W

’ (600)+W

Z

Z’ (300)

Z’ (600)

tt

H (120)→gg

H (160)→gg

H (250)→gg

t (t-channel)

VBF H (120)

VBF H (160)

VBF H (250)

LHC 7 TeV

CT10CTEQ6.6CT10W

0.7 0.8 0.9 1 1.1 1.2 1.3

+H

-HW

+HW

HZ

t (s-channel)

+W

’ (300)+W

’ (600)+W

Z

Z’ (300)

Z’ (600)

tt

H (120)→gg

H (160)→gg

H (250)→gg

t (t-channel)

VBF H (120)

VBF H (160)

VBF H (250)

Tevatron

CT10CTEQ6.6

CT10W


Backup slides


S-ACOT input parameters

At Q ≈ mc, F c2 depends significantly on

1. Charm mass: mc = 1.3 GeV in CT10

2. Factorization scale: µ =√

Q2 + κm2c ; κ = 1 in CT10

3. Rescaling variable ζ(λ) for matching in γ∗c channels(Tung et al., hep-ph/0110247; Nadolsky, Tung, PRD79, 113014 (2009))

Fi(x,Q2) =∑

a,b

∫ 1

ζ

dξ

ξfa(ξ, µ) Ca

b,λ

(ζ

ξ,Q

µ,mi

µ

)

x = ζ/(

1 + ζλ · (4m2c)/Q2

), with 0 ≤ λ . 1

CT10 usesζ(0) ≡ χ ≡ x

(1 + 4m2

c/Q2),

motivated by momentum conservation

1+ ��Mf

2

Q2 Ζ=ΧACOT HΛ=0L

Ζ=x HΛ®¥L

Λ=0.1

Λ=0.2

Λ=1 Phy

sica

lthr

esho

ld:W=

Mf;Ζ=

1

10-4 0.001 0.01 0.1 1

1

x

Res

calin

gfa

ctorΖ�x


CT10(W): radiative contributions to A`(y`)

¥ Default calculation: A`(y) at NNLL-NLO, using lookup tablesfor σ(p`

T , y`)NNLL+NLO/σ(p`T , y`)LO from ResBos [Balazs, Yuan, PRD 56, 5558

(1997); Landry, Brock, P.N. Yuan, PRD67, 073016 (2003)].

¥ Cross check: include NNLO corrections at QT ≈ MW [Arnold &

Reno, 1989]; A`(y`) changes by a few percent at the highest y`

and pT > 35 GeV

I magnitude of changes is comparable with full NNLO terms(Catani, Ferrera, and Grazzini, JHEP 05, 006 (2010))

I changes are small compared to the experimental errors


CT10 and CT10W predictions for Ae(ye) (D0 Run-2)

ey

0 0.5 1 1.5 2 2.5 3

) e (y e

A

-0.8

-0.6

-0.4

-0.2

0

0.2

> 25 GeVeT

p

=1.96 TeVS+X ν ± e→ ± W→pp

> 25 GeVνTE

)-1D0 electron data (0.75 fbCT10 (Solid band)CTEQ6.6 (Hatched band)

ey

0 0.5 1 1.5 2 2.5 3

) e (y e

A

-0.8

-0.6

-0.4

-0.2

0

0.2

< 35 GeVeT

25 GeV < p

=1.96 TeVS+X ν ± e→ ± W→pp

> 25 GeVνTE


ey

0 0.5 1 1.5 2 2.5 3

) e (y e

A

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

> 35 GeVeT

p

=1.96 TeVS+X ν ± e→ ± W→pp

> 25 GeVνTE


ey

0 0.5 1 1.5 2 2.5 3

) e (y e

A

-0.8

-0.6

-0.4

-0.2

0

0.2

> 25 GeVeT

p

=1.96 TeVS+X ν ± e→ ± W→pp

> 25 GeVνTE


ey

0 0.5 1 1.5 2 2.5 3

) e (y e

A

-0.8

-0.6

-0.4

-0.2

0

0.2

< 35 GeVeT

25 GeV < p

=1.96 TeVS+X ν ± e→ ± W→pp > 25 GeVν

TE


ey

0 0.5 1 1.5 2 2.5 3

) e (y e

A

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

> 35 GeVeT

p

=1.96 TeVS+X ν ± e→ ± W→pp

> 25 GeVνTE



Do CTEQ PDFs disagree with D0 (di)-jet data?Pumplin et al., PRD 80 (2009) 014019: no significant tension betweenCTEQ PDFs and incl. jet data; D0 presentation exaggerates the“discrepancy”

Data and NLO theory, from the D0 paper and CT09 analysis

D0 Coll., arXiv:0802.2400 (700 pb−1)

0.5

1.0

1.5

|y| < 0.4

DØ Run II-1L = 0.70 fb

= 0.7coneR

50 100 200 3000.0

0.5

1.0

1.5

1.2 < |y| < 1.6

NLO scale uncertainty

0.4 < |y| < 0.8

T = p

Fµ =

RµNLO pQCD

+non-perturbative corrections

50 100 200 300

1.6 < |y| < 2.0

CTEQ6.5M with uncertainties

MRST2004

0.8 < |y| < 1.2

DataSystematic uncertainty

50 100 200 300

2.0 < |y| < 2.4

(GeV)T

p

data

/ th

eory

(GeV)T

p

data

/ th

eory

(GeV)T

p

data

/ th

eory

(GeV)T

p

data

/ th

eory

(GeV)T

p

data

/ th

eory

(GeV)T

p

data

/ th

eory

“Discrepancy”?

(Shifted D0-Run 2 data)/CT09

Good agreement


Jet production: issues to consider

¥ Significant scaledependence

I Comparisons to CT10 PDFsmust use µ = pjet

T /2 , thesame scale as in the CT10fit

¥ Differences between NLOcodes; sensitivity toresummation of jetdifferential distributions(Alioli et al., arXiv:1012.3380)

(GeV)JJM200 400 600 800 1000 1200 1400

JJ

/dM

σR

atio

of

d

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

< 0.4max

|y|


CTEQ6.6 / CT10.00

CT10 / CT10.00

CT10W / CT10.00

(GeV)JJM200 400 600 800 1000 1200 1400

JJ

/dM

σR

atio

of

d

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

< 2.0max

1.6 < |y|


CTEQ6.6 / CT10.00

CT10 / CT10.00

CT10W / CT10.00

¥ Correlated systematic shifts reconcile the data with a widerrange of PDFs than in the standalone experimental analysis


Resummation effects in inclusive (di)jetproduction


Practical evaluation of the combined PDF+αs uncertainty

Several prescriptions of varying complexity for combining the PDFand αs uncertainties exist

In arXiv:1004.4624, we show that addition of the αs and PDFuncertainties in quadrature is entirely adequate in most practicalsituations

TheoremIn the quadratic approximation, the total αs+PDF uncertainty ∆σ forthe CT10 set, for αs(MZ) = 0.118± 0.002, is obtained by

∆X =√

∆X2CT10 + ∆X2

αs,

where¥ ∆XCT10 is the CTEQ6.6 PDF uncertainty from 44 PDFs with the same

αs(MZ) = 0.118

¥ ∆Xαs= (X0.120 −X0.116)/2 is the αs uncertainty computed with two

central CTEQ6.6AS PDFs for αs(MZ) = 0.116 and 0.120


Quadrature addition reproduces the exact PDF+αs uncertainty

Total PDF+αs errors ∆X are the same when found (a) from a fullfit with floating αs, or (b) by adding ∆XPDF and ∆Xαs inquadrature

10-5 10-4 10-3 0.01 0.02 0.05 0.1 0.2 0.5 0.7

0.6

0.8

1.0

1.2

1.4

x

CT

EQ

6.6A

S/C

TE

Q6.

6M

g at Q=2 GeV

10-5 10-4 10-3 0.01 0.02 0.05 0.1 0.2 0.5 0.7

0.6

0.8

1.0

1.2

1.4

x

CT

EQ

6.6A

S/C

TE

Q6.

6M

c at Q=2 GeV

¥ black – CTEQ6.6 PDFuncertainty

¥ Blue filled – PDF+αs

uncertainty of the fit withfloating αs(MZ)

¥ Green hatched – PDF+αs

uncertainty added inquadrature


news from cteq-tea pdf analysis · ¥ nnlo cteq global analysis (in progress) i validation of o(α2...

Documents