MadGraph5_aMC@NLO

tutorial

Pavia

!
Marco Zaro

LPTHE - Université Pierre et Marie Curie

Paris

1
 Part 1: 

getting familiar with

Marco Zaro, 12/14-05-2015 2

 What is

MadGraph5_aMC@NLO?
• It is an automatic meta-code that write the code for
computing the cross-section and generating events for
any process at colliders

• All the details are in arXiv:1405.0301

• NLO QCD corrections can be included

• Matrix elements of different multiplicities can be
combined

• at LO (CKKW or MLM)

• at NLO (FxFx or UNLOPS)

Marco Zaro, 12/14-05-2015 3

 Software prerequisites:

• Python 2.6 or 2.7

• Fortran compiler supporting quadruple precision
(needed for NLO)

• gfortran v4.6+ OK

• Optional:

• gnuplot

• FastJet (FJcore is included in the tarball)

• LHAPDF

• Herwig++ / Pythia8

Marco Zaro, 12/14-05-2015 4

 Where do I get it?
• On LaunchPad: https://launchpad.net/mg5amcnlo

Marco Zaro, 12/14-05-2015 5

 Where do I get it?
• On LaunchPad: https://launchpad.net/mg5amcnlo

•tar -xzf MG5_aMC_v2.3.0.tar.gz!
•cd MG5_aMC_v2_3_0!
•./bin/mg5_aMC

Marco Zaro, 12/14-05-2015 6

 Let’s start the tutorial
• On LaunchPad: https://launchpad.net/mg5amcnlo

•tar -xzf MG5_aMC_v2.3.0.tar.gz!
•cd MG5_aMC_v2_3_0!
•./bin/mg5_aMC!
•> tutorial

Marco Zaro, 12/14-05-2015 7

 Exercise 1:

Top pair production at LO
• Basic questions:

• Generate the process (following the tutorial)

• Which partonic subprocesses contribute?

• How many Feynman diagrams has each subprocess?

• Output the code

• Compute the cross-section at the LHC (8 TeV) for mt=170 GeV

• Extra questions:

• Are b-quarks included in the initial state? If not, how can I include
them?

• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change? What is that
‘WEIGHTED’?

• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV
Marco Zaro, 12/14-05-2015 8
 Exercise 1:

solution
• Generate the process (following the tutorial)

!
•> generate p p > t t~!
INFO: Checking for minimal orders which gives processes.
INFO: Please specify coupling orders to bypass this step.
INFO: Trying coupling order WEIGHTED=2
INFO: Trying process: g g > t t~ WEIGHTED=2
INFO: Process has 3 diagrams
INFO: Trying process: u u~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Trying process: u c~ > t t~ WEIGHTED=2
INFO: Trying process: c u~ > t t~ WEIGHTED=2
INFO: Trying process: c c~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Trying process: d d~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Trying process: d s~ > t t~ WEIGHTED=2
INFO: Trying process: s d~ > t t~ WEIGHTED=2
INFO: Trying process: s s~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Process u~ u > t t~ added to mirror process u u~ > t t~
INFO: Process c~ c > t t~ added to mirror process c c~ > t t~
INFO: Process d~ d > t t~ added to mirror process d d~ > t t~
INFO: Process s~ s > t t~ added to mirror process s s~ > t t~
5 processes with 7 diagrams generated in 0.075 s
Total: 5 processes with 7 diagrams

Marco Zaro, 12/14-05-2015 9

 Exercise 1:

solution
• Which partonic subprocesses contribute?

•> display processes!
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2

Marco Zaro, 12/14-05-2015 10

 Exercise 1:

solution
• Which partonic subprocesses contribute?

•> display processes!
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
u u~ > t t~ WEIGHTED=2
g g > t t~ WEIGHTED=2
page 1/1
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2
2 4 2 2
4
g
u~ t~ g t~

g g

t t
u g

3 1 3 1
1

diagram 1 QCD=2, QED=0 diagram 1 QCD=2, QED=0 diag

Marco Zaro, 12/14-05-2015 10 2 3

 Exercise 1:

solution
• Which partonic subprocesses contribute?

•> display processes!
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2

QCD
Z master formula:
X
(pp ! tt̄) = dx1 dx2 fa (x1 , µF )fb (x2 , µF ) ⇥ ˆ (ab ! tt̄)
ab

Marco Zaro, 12/14-05-2015 10

 What does it mean?
X Z
(pp ! tt̄) = dx1 dx2 fa (x1 , µF )fb (x2 , µF ) ⇥ ˆ (ab ! tt̄)
ab
• What is the probability to

u u~ > t t~ WEIGHTED=2
find parton a inside the P P

2
proton with momentum

u
fraction x? fa(x)
6.5 TeV 6.5 TeV

u~
diagram 1
• μF is a scale which separates
low energy from high

g
energy dynamics

• The partonic scattering QCD=2, QED=0

occurs at a reduced energy:

t~
t
ŝ = x1 x2 S = x1 x2 (13TeV)2
3

4
Marco Zaro, 12/14-05-2015 11
 Exercise 1:

solution
• How many Feynman diagrams has each subprocess?
u u~ > t t~ WEIGHTED=2

•> display diagrams

g g > t t~ WEIGHTED=2 page 1/1

2 2 4
4 2
g t~ 4

g t~
u~ t~
g
t~
g
t
g

g t
t
u
1 3 1 3

diagram 1 QCD=2, QED=0 diagram 2 QCD=2, QED=0

1 3

2 3 diagram 1 QCD=2, QED=0

g t

t
gg: 3 diagrams qq:̄ 1 diagram
g t~

1 4

diagram 3 QCD=2, QED=0

Marco Zaro, 12/14-05-2015 12

 Exercise 1:

solution
• Output the code

•> output mytestdir!
INFO: initialize a new directory: mytestdir
INFO: remove old information in mytestdir
INFO: Creating files in directory P0_gg_ttx
INFO: Generating Feynman diagrams for Process: g g > t t~ WEIGHTED=2
INFO: Finding symmetric diagrams for subprocess group gg_ttx
INFO: Creating files in directory P0_qq_ttx
INFO: Generating Feynman diagrams for Process: u u~ > t t~ WEIGHTED=2
INFO: Finding symmetric diagrams for subprocess group qq_ttx
History written to /Users/marcozaro/Physics/MadGraph/MG5_aMC_v2_2_2/mytestdir/Cards/proc_card_mg5.dat
Generated helas calls for 2 subprocesses (0 diagrams) in 0.000 s
Wrote files for 16 helas calls in 0.102 s
Export UFO model to MG4 format
ALOHA: aloha creates FFV1 routines
ALOHA: aloha creates VVV1 set of routines with options: P0
save configuration file to /Users/marcozaro/Physics/MadGraph/MG5_aMC_v2_2_2/mytestdir/Cards/me5_configuration.txt
INFO: Use Fortran compiler gfortran
INFO: Generate jpeg diagrams
INFO: Generate web pages
Output to directory /Users/marcozaro/Physics/MadGraph/MG5_aMC_v2_2_2/mytestdir done.

Marco Zaro, 12/14-05-2015 13

 Exercise 1:

solution
• Compute the cross-section at the LHC (8 TeV) for mt=170 GeV

•> launch!
The following switches determine which programs are run:
1 Run the pythia shower/hadronization: pythia=NOT INSTALLED
2 Run PGS as detector simulator: pgs=NOT INSTALLED
3 Run Delphes as detector simulator: delphes=NOT INSTALLED
4 Decay particles with the MadSpin module: madspin=OFF
5 Add weight to events based on coupling parameters: reweight=OFF
Either type the switch number (1 to 5) to change its default setting,
or set any switch explicitly (e.g. type 'madspin=ON' at the prompt)
Type '0', 'auto', 'done' or just press enter when you are done.
[0, 4, 5, auto, done, madspin=ON, madspin=OFF, madspin, reweight=ON, ... ][60s to answer]

•> 0 (let’s keep it simple ;-)

Do you want to edit a card (press enter to bypass editing)?
1 / param : param_card.dat
2 / run : run_card.dat
you can also
- enter the path to a valid card or banner.
- use the 'set' command to modify a parameter directly.
The set option works only for param_card and run_card.
Type 'help set' for more information on this command.
- call an external program (ASperGE/MadWidth/...).
Type 'help' for the list of available command
[0, done, 1, param, 2, run, enter path][60s to answer]

•! edit the cards

Marco Zaro, 12/14-05-2015 14
 Exercise 1:

solution
• !Compute the cross-section at the LHC (8 TeV) for mt=170 GeV

run_card
param_card

Marco Zaro, 12/14-05-2015 15

 Exercise 1:

solution
• Compute the cross-section at the LHC (8 TeV) for mt=172 GeV

• One can also set the parameters without editing the cards
(useful for scripting)

•> set ebeam1 4000!
•> set ebeam2 4000!
•> set MT 172.!
•> done

Marco Zaro, 12/14-05-2015 16

 Exercise 1:

solution
• Compute the cross-section at the LHC (8 TeV) for mt=172 GeV

• One can also set the parameters
!
without editing the cards
(useful for scripting)
. . .
Working on SubProcesses

•> set ebeam1 4000! P0_gg_ttx

P0_qq_ttx
INFO: Idle: 0, Running: 1, Completed: 1 [ current time: 15h13 ]

•> set ebeam2 4000! INFO: End survey

refine 10000
Creating Jobs

•> set MT 172.! INFO: Refine results to 10000

P0_gg_ttx

•> done

P0_qq_ttx
INFO: Idle: 6, Running: 4, Completed: 3 [ 3.2s ]
INFO: Idle: 2, Running: 4, Completed: 7 [ 6.6s ]
INFO: Idle: 0, Running: 1, Completed: 12 [ 9.7s ]
INFO: Combining runs
INFO: finish refine
refine 10000
Creating Jobs
INFO: Refine results to 10000
P0_gg_ttx
P0_qq_ttx
INFO: Combining runs
INFO: finish refine
combine_events
INFO: Combining Events
=== Results Summary for run: run_01 tag: tag_1 ===
!
Cross-section : 160.1 +- 0.2302 pb
Nb of events : 10000

Marco Zaro, 12/14-05-2015 16

 Monitor via the web interface

Marco Zaro, 12/14-05-2015 17

 Exercise 1:

solution
• Script it:

• open a text file (mymg5amc.txt) and put the commands
inside:

generate p p > t t~!
output mytestdir!
launch!
set ebeam1 4000!
set ebeam2 4000!
set MT 172!
• launch MG5_aMC@NLO with that file

•./bin/mg5_amc mymg5amc.txt
Marco Zaro, 12/14-05-2015 18
 Intermezzo:

From Feynman diagrams to the amplitude u u~ > t t~ WEIGHTED=2

• What is the amplitude for uu→

̄ t t?

̄
• Peskin & Schroeder answer:

2 4

4
g u~ t~
|M!|2 = T r[p/2 γ µ p/1 γ ν ] 2
T r[( p
/ 3 + m)γ µ
( p
/ 4 − m)γ ν
] = ...
(p1 · p2 )
• Gives a clean and compact expression
 g

(for simple processes)

u
t

• Number of terms ~ N diag

2

1 3

• OK for simple processes, not for complex ones! diagram 1 QCD=2, QED=0

Marco Zaro, 12/14-05-2015 19

 Intermezzo:

From Feynman diagrams to the amplitude u u~ > t t~ WEIGHTED=2

•
|M |2 What
= T r[p/2 γis
µ the
p/1 γ ν ] amplitude
g 4
T r[( p
/3 for
+ uu
m)γ µ →t t?

( ̄
p /4 − ̄
m)γ ν
]2 = . . .
• Helicity-based formula
g
(p1 · p2 ) 2 4

µν u~ t~
2
|M |! s1 s2 s3 s4
= g v̄(p2 ) γµ u(p1 )
s2 s1
ū(p3 ) γν u(p4 )
s3 s4
p1 · p2
! ! 2
g

2
|M | = (|M |s1 s2 s3 s4 ) t
! u

• Much simpler expression

s1 s2 s3 s4

1 3

• Number of terms ~ Ndiag diagram 1 QCD=2, QED=0

• Suitable for numeric codes!

Marco Zaro, 12/14-05-2015 20

 Intermezzo:

From Feynman diagrams to the amplitude u u~ > t t~ WEIGHTED=2

•
|M |2 What
= T r[p/2 γis
µ the
p/1 γ ν ] amplitude
g 4
T r[( p
/3 for
+ uu
m)γ µ →t t?

( ̄
p /4 − ̄
m)γ ν
] = 2. . .
• Helicity-based formulag
(p1 · p2 ) 2 4

µν u~ t~
2
|M | s1 s2 s3 s4
= g v̄(p2 ) γµ u(p1 )
s2 s1
ū(p3 ) γν u(p4 )
s3 s4
p1 · p2 g
! 2
2
|M | = (|M |s1 s2 s3 s4 ) t
u
---------- s1 s2 s3 s4
BEGIN CODE
3
---------- 1
CALL IXXXXX(P(0,1),ZERO,NHEL(1),+1*IC(1),W(1,1))
diagram 1 QCD=2, QED=0
CALL OXXXXX(P(0,2),ZERO,NHEL(2),-1*IC(2),W(1,2))
CALL OXXXXX(P(0,3),MDL_MT,NHEL(3),+1*IC(3),W(1,3))
CALL IXXXXX(P(0,4),MDL_MT,NHEL(4),-1*IC(4),W(1,4))
CALL FFV1P0_3(W(1,1),W(1,2),GC_11,ZERO,ZERO,W(1,5))
Amplitude(s) for diagram number 1
CALL FFV1_0(W(1,4),W(1,3),W(1,5),GC_11,AMP(1))

Marco Zaro, 12/14-05-2015 21

 Intermezzo:

From Feynman diagrams to the amplitude u u~ > t t~ WEIGHTED=2

•
|M |2 What
= T r[p/2 γis
µ the
p/1 γ ν ] amplitude
g 4
T r[( p
/3 for
+ uu
m)γ µ →t t?

( ̄
p /4 − ̄
m)γ ν
] = 2. . .
• Helicity-based formulag
(p1 · p2 ) 2
W(j,2) W(j,4)
4

µν u~ t~
2
|M | s1 s2 s3 s4
= g v̄(p2 ) γµ u(p1 )
s2 s1
ū(p3 ) γν u(p4 )
s3 s4
p1 · p2 g
! 2
2
|M | = (|M |s1 s2 s3 s4 ) t
u
---------- s1 s2 s3 s4
BEGIN CODE W(j,1)
---------- 1 W(j,3) 3

CALL IXXXXX(P(0,1),ZERO,NHEL(1),+1*IC(1),W(1,1))
diagram 1 QCD=2, QED=0
CALL OXXXXX(P(0,2),ZERO,NHEL(2),-1*IC(2),W(1,2))
CALL OXXXXX(P(0,3),MDL_MT,NHEL(3),+1*IC(3),W(1,3))
CALL IXXXXX(P(0,4),MDL_MT,NHEL(4),-1*IC(4),W(1,4))
CALL FFV1P0_3(W(1,1),W(1,2),GC_11,ZERO,ZERO,W(1,5))
Amplitude(s) for diagram number 1
CALL FFV1_0(W(1,4),W(1,3),W(1,5),GC_11,AMP(1))

Marco Zaro, 12/14-05-2015 21

 Intermezzo:

From Feynman diagrams to the amplitude u u~ > t t~ WEIGHTED=2

•
|M |2 What
= T r[p/2 γis
µ the
p/1 γ ν ] amplitude
g 4
T r[( p
/3 for
+ uu
m)γ µ →t t?

( ̄
p /4 − ̄
m)γ ν
] = 2. . .
• Helicity-based formulag
(p1 · p2 ) 2
W(j,2) W(j,4)
4

µν u~ t~
2
|M | s1 s2 s3 s4
= g v̄(p2 ) γµ u(p1 )
s2 s1
ū(p3 ) γν u(p4 )
s3 s4
W(j,5)
p1 · p2 g
! 2
2
|M | = (|M |s1 s2 s3 s4 ) t
u
---------- s1 s2 s3 s4
BEGIN CODE W(j,1)
---------- 1 W(j,3) 3

CALL IXXXXX(P(0,1),ZERO,NHEL(1),+1*IC(1),W(1,1))
diagram 1 QCD=2, QED=0
CALL OXXXXX(P(0,2),ZERO,NHEL(2),-1*IC(2),W(1,2))
CALL OXXXXX(P(0,3),MDL_MT,NHEL(3),+1*IC(3),W(1,3))
CALL IXXXXX(P(0,4),MDL_MT,NHEL(4),-1*IC(4),W(1,4))
CALL FFV1P0_3(W(1,1),W(1,2),GC_11,ZERO,ZERO,W(1,5))
Amplitude(s) for diagram number 1
CALL FFV1_0(W(1,4),W(1,3),W(1,5),GC_11,AMP(1))

Marco Zaro, 12/14-05-2015 21

 Intermezzo:

From Feynman diagrams to the amplitude u u~ > t t~ WEIGHTED=2

•
|M |2 What
= T r[p/2 γis
µ the
p/1 γ ν ] amplitude
g 4
T r[( p
/3 for
+ uu
m)γ µ →t t?

( ̄
p /4 − ̄
m)γ ν
] = 2. . .
• Helicity-based formulag
(p1 · p2 ) 2
W(j,2) W(j,4)
4

µν u~ t~
2
|M | s1 s2 s3 s4
= g v̄(p2 ) γµ u(p1 )
s2 s1
ū(p3 ) γν u(p4 )
s3 s4
W(j,5)
p1 · p2 g
! 2
2
|M | = (|M |s1 s2 s3 s4 ) t
u
---------- s1 s2 s3 s4
BEGIN CODE W(j,1)
---------- 1 W(j,3) 3

CALL IXXXXX(P(0,1),ZERO,NHEL(1),+1*IC(1),W(1,1))
diagram 1 QCD=2, QED=0
CALL OXXXXX(P(0,2),ZERO,NHEL(2),-1*IC(2),W(1,2))
CALL OXXXXX(P(0,3),MDL_MT,NHEL(3),+1*IC(3),W(1,3))
CALL IXXXXX(P(0,4),MDL_MT,NHEL(4),-1*IC(4),W(1,4)) AMP(1)
CALL FFV1P0_3(W(1,1),W(1,2),GC_11,ZERO,ZERO,W(1,5))
Amplitude(s) for diagram number 1
CALL FFV1_0(W(1,4),W(1,3),W(1,5),GC_11,AMP(1))

Marco Zaro, 12/14-05-2015 21

 Exercise 1:

Extra questions:
• Are b-quarks included in the initial state? If not, how can I
include them?

•> display processes!
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2

• No b-quark appears. Note that at the startup you have

Defined multiparticle p = g u c d s u~ c~ d~ s~
Defined multiparticle j = g u c d s u~ c~ d~ s~

• You can add the b/b̄ to the multiparticle labels

•> define p = p b b~!
Defined multiparticle p = g u c d s u~ c~ d~ s~ b b~

•> display multiparticles!

• For consistency one should use a model with mb=0

•> import model sm-no_b_mass!
Marco Zaro, 12/14-05-2015 22
 Exercise 1:

Extra questions:

Marco Zaro, 12/14-05-2015 23

 Exercise 1:

Extra questions:
• Are b-quarks included in the initial state? If not, how can I
include them?

Marco Zaro, 12/14-05-2015 23

 Exercise 1:

Extra questions:
• Are b-quarks included in the initial state? If not, how can I
include them?
• Regenerate the process

•> generate p p > t t~!
•> display processes
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2
Process: b b~ > t t~ WEIGHTED=2

Marco Zaro, 12/14-05-2015 23

 Exercise 1:

Extra questions:
• Are b-quarks included in the initial state? If not, how can I
include them?
• Regenerate the process

•> generate p p > t t~!
•> display processes
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2
Process: b b~ > t t~ WEIGHTED=2

• Does it make a big difference?

•> output!
•> launch!
•> set ebeam1 4000!
•> set ebeam2 4000!
•> set MT 172
Marco Zaro, 12/14-05-2015 23
 Exercise 1:

Extra questions:
• Are b-quarks included in the initial state? If not, how can I
include them?
• Regenerate the process

•> generate p p > t t~!
•> display processes
Process: g g > t t~ WEIGHTED=2
Process: u u~ > t t~ WEIGHTED=2
Process: c c~ > t t~ WEIGHTED=2
Process: d d~ > t t~ WEIGHTED=2
Process: s s~ > t t~ WEIGHTED=2
Process: b b~ > t t~ WEIGHTED=2

• Does it make a big difference?

•> output! Cross-section : 160.4 +- 0.231 pb
Nb of events : 10000
• > launch! !


Without b

•> set ebeam1 4000! Cross-section : 160.1 +- 0.2302 pb
Nb of events : 10000
•> set ebeam2 4000!
•> set MT 172
Marco Zaro, 12/14-05-2015 23
 Exercise 1:

Extra questions:

Marco Zaro, 12/14-05-2015 24

 Exercise 1:

Extra questions:
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change? What is that
‘WEIGHTED’?

Marco Zaro, 12/14-05-2015 24

 Exercise 1:

Extra questions:
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change? What is that
‘WEIGHTED’? u u~ > t t~ WEIGHTED=2
g g > t t~ WEIGHTED=2
page 1/1

•> display diagrams 2 4 2 4 2

g
u~ t~ g t~

g g

t t
u g

3 1 3 1
1

• No photon/z appear. diagram 1 QCD=2, QED=0 diagram 1 QCD=2, QED=0 diagram 2

• Are we missing anything important? 2

g t
3

Marco Zaro, 12/14-05-2015 24 t

 Exercise 1:

Extra questions:
!
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change? What is that
‘WEIGHTED’?

•> display diagrams!
• No photon/z appear.

• Are we missing anything important?

Marco Zaro, 12/14-05-2015 25

 Exercise 4 1:
4
µg g
r[ 2 µ
Mp/|2 γ=p/T
1 γr[p
ν
]/2 γ pExtra
/1 γ 2] Tquestions:
ν
r[(p/3 +2 Tm)γ
r[(p/3(p+
µ
/4 − m)γ(p/4] =
m)γ µ ν
!
(p1 · p2 ) (p1 · p2 )
• Are diagrams with photons/z included?µνIf not, how can I include
them? How
2 much does
2 the g
cross-section change? gµν
What is that
s s
|M
2 3 4
=| g v̄(p2 )= gγµv̄(p
s s s
1 2 3 4s
‘WEIGHTED’?

s 2
u(p21)) γµ u(p1 )ū(p3 ) γνū(p
ss21 s 1 s 3
u(p34))ss

• > display diagrams! p 1 · p 2 p 1 · p 2

• No photon/z !
appear.
!
2 s1 s2 s3 s4 2 s1 s2 s3 s4 2
• Are|Mwe |missing
= anything
u u~ > t t~ WEIGHTED=2 2
=(|M |
|M | important?
(|M) | )
u u~ > t t~ QED=2
page 1/1
page 1/1

2 4 2s1 s2 s3 s42 s1 s2 s3 s4 4
2 2 4

u~ t~ u~ u~
t~ t~

2 22 2
g
∼ αs ∼∼ααs e a
∼ αe g

t t t
u u u

1 3 3 3
1 1

diagram 1 QCD=2, QED=0 diagram 1 QCD=0, QED=2 diagram 2 QCD=2, QED=0

Marco Zaro, 12/14-05-2015 25

2 4
 4
2 g
|M 4| = T r[p/24γ p/1 γ ] Exercise 1:
T r[( p
µ ν
2 /3
g g (p 1 · p 2 )
r[ 2 µ
Mp/|2 γ=p/T1 γr[p
ν
]/2 γ p/1 γ 2] T r[(p/3 +2 T
µ ν
Extra questions:
m)γ
r[(p/3(p+
µ
/4 − m)γ m)γ(p/4] =
µ ν
!
(p 1 · p 2 ) (p1 · p2 )
2
• |M | = gIf not,
v̄(phow
2 ) canγµI u(p 1)
s1 s2 s3 s4 s2 s1
Are diagrams with photons/z included? include
them? How much does the g µν
cross-section change? g µν
What is that
s s
|M
2 3 4 s s2
=| g v̄(p2 )= gγµv̄(p
s
1 2 3 4s
‘WEIGHTED’?

s 2 2
u(p21)) γµ u(p1 )ū(p3 ) !
ss21 s 1 s 3
γνū(p
u(p34))ss

•
> display diagrams! p · p
1 |M2| =p1 · p2
2
(|M |
• No photon/z !
appear.
!
2 s1 s2 s3 s4 2 s1s1ss22ss33ss44 2
• Are|M
we |missing
u u~ > t t~ WEIGHTED=2
= anything 2
=(|M |
|M | important?
(|M) |u u~ > t t~ QED=2
) page 1/1
page 1/1

2
2 4 2s1 s2 s3 s42 s1 s2 s3 s4 4
2 ∼ 2 αs ∼ 4

αe
u~ t~ u~ u~
t~ t~

2 22 2
g
∼ αs ∼∼ααs e a
∼ αe ! 0.1g

u
t
u
t
u αs t

1 3 3 3
1 1

diagram 1 QCD=2, QED=0 diagram 1 QCD=0, QED=2 diagram 2 QCD=2, QED=0

Marco Zaro, 12/14-05-2015 25

2 4
 Exercise 1:

Extra questions:
!
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change? 

What is that ‘WEIGHTED’?

•> display diagrams!
• No photon/z appear.

• Are we missing anything important? Does not seem the case

• How to have them anyway?

• MG5 exploits the hierarchy between QCD and QED
couplings in order to give the leading (i.e. with most QCD)
contribution to the cross-section by default

• It assign WEIGHTED order =1(=2) to QCD (QED) vertices
and generates the process with minimum WEIGHTED order
Marco Zaro, 12/14-05-2015 26
 Exercise 1:

Extra questions:
!
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change? 

What is that ‘WEIGHTED’?

•> display diagrams!
• No photon/z appear.

• Are we missing anything important? Does not seem the case

• How to have them anyway?

• MG5 exploits the hierarchy between QCD and QED
couplings in order to give the leading (i.e. with most QCD)
contribution to the cross-section by default

• It assign WEIGHTED order =1(=2) to QCD (QED) vertices
and generates the process with minimum WEIGHTED order
Marco Zaro, 12/14-05-2015 26
 Exercise 1:

Extra questions:

Marco Zaro, 12/14-05-2015 27

 Exercise 1:

Extra questions:
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change?
•> generate p p > t t~ WEIGHTED=4
•> display diagrams

Marco Zaro, 12/14-05-2015 27

 Exercise 1:

Extra questions:
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change?
s s~ > t t~ WEIGHTED=4 page 1/

•> generate p p > t t~ WEIGHTED=4

•> display diagrams 2

s~ t~
4 2

s~ t~
4

a g

t t
s s

1 3 1 3

diagram 1 QCD=0, QED=2 diagram 2 QCD=2, QED=0

2 4

s~ t~

t
s

1 3

Marco Zaro, 12/14-05-2015 27 diagram 3 QCD=0, QED=2

 Exercise 1:

Extra questions:
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change?
s s~ > t t~ WEIGHTED=4 page 1/

•> generate p p > t t~ WEIGHTED=4

•> display diagrams 2 4 2 4

•> output … s~ t~ s~ t~

•> launch
a g

t t
s s

•> … 1 3 1 3

diagram 1 QCD=0, QED=2 diagram 2 QCD=2, QED=0

2 4

s~ t~

t
s

1 3

Marco Zaro, 12/14-05-2015 27 diagram 3 QCD=0, QED=2

 Exercise 1:

Extra questions:
• Are diagrams with photons/z included? If not, how can I include
them? How much does the cross-section change?
s s~ > t t~ WEIGHTED=4 page 1/

•> generate p p > t t~ WEIGHTED=4

•> display diagrams 2 4 2 4

•> output … s~ t~ s~ t~

•> launch
a g

t t
s s

•> … 1 3 1 3

diagram 1 QCD=0, QED=2 diagram 2 QCD=2, QED=0

Cross-section : 160.8 +- 0.1999 pb 2 4

Nb of events : 10000
! s~ t~

WEIGHTED=2
z

Cross-section : 160.4 +- 0.231 pb t

Nb of events : 10000
s

1 3

Marco Zaro, 12/14-05-2015 27 diagram 3 QCD=0, QED=2

 Exercise 1:

Extra questions:

Marco Zaro, 12/14-05-2015 28

 Exercise 1:

Extra questions:
• Alternatively, one can specify the coupling powers

•> generate p p > t t~ QED=2!
• orders which are not specified are unconstrained

Marco Zaro, 12/14-05-2015 28

 Exercise 1:

Extra questions:
• Alternatively, one can specify the coupling powers

•> generate p p > t t~ QED=2!
• orders which are not specified are unconstrained
• In order to have only the QED contribution

•> generate p p > t t~ QED=2 QCD=0

Marco Zaro, 12/14-05-2015 28

 Exercise 1:

Extra questions:
!
• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV

• Be smart! Script it!

• Create a txt file myttbar_scan.txt

generate p p > t t~
output mytestdir2
launch
set ebeam1 4000
set ebeam2 4000
set MT 170
launch
set MT 172
launch
set MT 174
launch
set MT 176
launch
set MT 178
launch
set MT 180

•./bin/mg5_aMC myttbar_scan.txt!
Marco Zaro, 12/14-05-2015 29
 Exercise 1:

Extra questions:
!
• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV

• Be smart! Script it!

• You can also launch an existing folder, without regenerating the
code

launch mytestdir2
set ebeam1 4000
set ebeam2 4000
set MT 170
launch
set MT 172
launch
set MT 174
launch
set MT 176
launch
set MT 178
launch
set MT 180

Marco Zaro, 12/14-05-2015 30

 Exercise 1:

Extra questions:
!
• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV

Marco Zaro, 12/14-05-2015 31

 Exercise 1:

Extra questions:
!
• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV

which folder is what?

Marco Zaro, 12/14-05-2015 31
 Exercise 1:

Extra questions:
!
• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV

• Be smart! Script it!

• You can specify the name (instead of run_01…) with -n NAME

launch mytestdir2 -n run_MT170
set ebeam1 4000
set ebeam2 4000
set MT 170
launch -n run_MT172
set MT 172
launch -n run_MT174
set MT 174
launch -n run_MT176
set MT 176
launch -n run_MT178
set MT 178
launch -n run_MT180
set MT 180

Marco Zaro, 12/14-05-2015 32

 Exercise 1:

Extra questions:
!
• Recompute the t t ̄ cross-section for mt=170, 172, 174 … 180 GeV

Marco Zaro, 12/14-05-2015 33

 Intermezzo:

The Standalone output mode
• Suppose you want just the matrix element for a given
process

• Why shall I need it?

• You want to cross-check one computation you did

• You have your own integrator, and you need to plug
the matrix element in

• The Standalone output mode is what you need

•> generate u u~ > t t~!
•> output standalone my_uux_ttx_SA!
•> launch

Marco Zaro, 12/14-05-2015 34

 The Standalone output mode
mdl_lam = 0.12886910601690263
mdl_yb = 2.6995554250465490E-002
mdl_yt = 0.99366614581500623
mdl_ytau = 1.0206617000654717E-002
mdl_muH = 88.388347648318430
mdl_I1x33 = ( 2.6995554250465490E-002, 0.0000000000000000 )
mdl_I2x33 = ( 0.99366614581500623 , 0.0000000000000000 )
mdl_I3x33 = ( 0.99366614581500623 , 0.0000000000000000 )
mdl_I4x33 = ( 2.6995554250465490E-002, 0.0000000000000000 )
mdl_ee__exp__2 = 9.4835522759998875E-002
mdl_sw__exp__2 = 0.22224648578577769
mdl_cw__exp__2 = 0.77775351421422245
Internal Params evaluated point by point
----------------------------------------

mdl_sqrt__aS = 0.34351128074635334
mdl_G__exp__2 = 1.4828317324943823
Couplings of sm
---------------------------------

GC_11 = 0.00000E+00 0.12177E+01

1000.0000000000000 500.00000000000000
!
Phase space point:
!
-----------------------------------------------------------------------------
n E px py pz m
1 0.5000000E+03 0.0000000E+00 0.0000000E+00 0.5000000E+03 0.0000000E+00
2
3
0.5000000E+03 0.0000000E+00 0.0000000E+00 -0.5000000E+03 0.0000000E+00
0.5000000E+03 0.1040730E+03 0.4173556E+03 -0.1872274E+03 0.1730000E+03 Momenta and masses
4 0.5000000E+03 -0.1040730E+03 -0.4173556E+03 0.1872274E+03 0.1730000E+03
-----------------------------------------------------------------------------
Matrix element = 0.61562818665255248 GeV^ 0
----------------------------------------------------------------------------- Matrix element value
Marco Zaro, 12/14-05-2015 35
 Exercise 2:

decay chains
• Theory: the top quark is an unstable particle:

• It decays: ~100% of times into b W

• The W boson decays too:

• 67% (2/3) of times into hadrons

• 22% (2/9) of times into “leptons” (e-νe or μ-νμ)

• 11% (1/9) of times into 𝝉-ν𝛕

• A decayed pair of top quarks can be classified as:

• hadronic (both tops to hadrons)

• semileptonic (one top to hadrons, the other to leptons)

• dileptonic (both quarks to leptons)
Marco Zaro, 12/14-05-2015 36
 Exercise 2:

decay chains
• Questions:

• How often a top pair decays hadronically/semi-leptonically/di-
leptonically?

• Learn the syntax to specify decay chains

• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1

• What is the difference with p p > l+ l- vl vl~ b b~?

Marco Zaro, 12/14-05-2015 37

 Exercise 2:

Solution

Marco Zaro, 12/14-05-2015 38

 Exercise 2:

Solution
• Questions:
• How often a top pair decays hadronically/semi-leptonically/di-
leptonically?
• Since the top always decays to Wb, 

look at how a pair of W decays 

(b’s are stable)

Marco Zaro, 12/14-05-2015 38

 Exercise 2:

Solution
• Questions:
• How often a top pair decays hadronically/semi-leptonically/di-
leptonically?
• Since the top always decays to Wb, 

look at how a pair of W decays 

(b’s are stable)
• Hadronically: 2/3 * 2/3 = 4/9

Marco Zaro, 12/14-05-2015 38

 Exercise 2:

Solution
• Questions:
• How often a top pair decays hadronically/semi-leptonically/di-
leptonically?
• Since the top always decays to Wb, 

look at how a pair of W decays 

(b’s are stable)
• Hadronically: 2/3 * 2/3 = 4/9

Marco Zaro, 12/14-05-2015 38

 Exercise 2:

Solution
• Questions:
• How often a top pair decays hadronically/semi-leptonically/di-
leptonically?
• Since the top always decays to Wb, 

look at how a pair of W decays 

(b’s are stable)
• Hadronically: 2/3 * 2/3 = 4/9
• Semi-lep. (incl. 𝝉): 2 * 1/3 * 2/3 = 4/9

Marco Zaro, 12/14-05-2015 38

 Exercise 2:

Solution
• Questions:
• How often a top pair decays hadronically/semi-leptonically/di-
leptonically?
• Since the top always decays to Wb, 

look at how a pair of W decays 

(b’s are stable)
• Hadronically: 2/3 * 2/3 = 4/9
• Semi-lep. (incl. 𝝉): 2 * 1/3 * 2/3 = 4/9
• Di-lep. (incl. 𝝉): 1/3 * 1/3 = 1/9

Marco Zaro, 12/14-05-2015 38

 Exercise 2:

Solution
• Questions:

• Learn the syntax to specify decay chains

• > help generate!
-- generate diagrams for a given process
General leading-order syntax:
o generate INITIAL STATE > REQ S-CHANNEL > FINAL STATE $ EXCL S-CHANNEL / FORBIDDEN PARTICLES COUP1=ORDER1 COUP2^2=ORDER2
@N
o Example: generate l+ vl > w+ > l+ vl a $ z / a h QED=3 QCD=0 @1
> Alternative required s-channels can be separated by "|":
b b~ > W+ W- | H+ H- > ta+ vt ta- vt~
> If no coupling orders are given, MG5 will try to determine
orders to ensure maximum number of QCD vertices.
> Desired coupling orders combination can be specified directly for
the squared matrix element by appending '^2' to the coupling name.
For example, 'p p > j j QED^2==2 QCD^==2' selects the QED-QCD
interference terms only. The other two operators '<=' and '>' are
supported. Finally, a negative value COUP^2==-I refers to the
N^(-I+1)LO term in the expansion of the COUP order.
> To generate a second process use the "add process" command
Decay chain syntax: Something like this!
o core process, decay1, (decay2, (decay2', ...)), ... etc
o Example: generate p p > t~ t QED=0, (t~ > W- b~, W- > l- vl~), t > j j b @2
> Note that identical particles will all be decayed

• > generate p p > t t~, (t > w+ b, w+ > l+ vl), 


(t~ > w- b~, w- > l- vl~)!
!
Marco Zaro, 12/14-05-2015 39
 Exercise 2:

Solution

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect?

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect?
• Something like 160 * 1/9 = 18 pb? Cross-section : 5.65 +- 0.01823 pb
Nb of events : 10000

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect?
• Something like 160 * 1/9 = 18 pb? Cross-section : 5.65 +- 0.01823 pb
Nb of events : 10000

• Wait: what is l+/l-?

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect?
• Something like 160 * 1/9 = 18 pb? Cross-section : 5.65 +- 0.01823 pb
Nb of events : 10000

• •Wait: what is l+/l-?

> display multi particles
Multiparticle
all = g u c d
l- = e- mu-
labels:
s u~ c~ d~ s~ a ve vm vt e- mu- ve~ vm~ vt~ e+ mu+ t

j = g u c d s u~ c~ d~ s~
vl = ve vm vt
l+ = e+ mu+
p = g u c d s u~ c~ d~ s~
vl~ = ve~ vm~ vt~

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect? 4/81 = 7.9

• Something like 160 * 1/9 = 18 pb? Cross-section : 5.65 +- 0.01823 pb
Nb of events : 10000

• •Wait: what is l+/l-?

> display multi particles
Multiparticle
all = g u c d
l- = e- mu-
labels:
s u~ c~ d~ s~ a ve vm vt e- mu- ve~ vm~ vt~ e+ mu+ t

j = g u c d s u~ c~ d~ s~
vl = ve vm vt
l+ = e+ mu+
p = g u c d s u~ c~ d~ s~
vl~ = ve~ vm~ vt~

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect? 4/81 = 7.9

• Something like 160 * 1/9 = 18 pb? Cross-section : 5.65 +- 0.01823 pb
Nb of events : 10000

• •Wait: what is l+/l-?

> display multi particles
Multiparticle
all = g u c d
l- = e- mu-
labels:
s u~ c~ d~ s~ a ve vm vt e- mu- ve~ vm~ vt~ e+ mu+ t

•
j = g u c d s u~ c~ d~ s~
Check the run_card… vl = ve vm vt
l+ = e+ mu+
p = g u c d s u~ c~ d~ s~
vl~ = ve~ vm~ vt~

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect? 4/81 = 7.9

• Something like 160 * 1/9 = 18 pb? Cross-section : 5.65 +- 0.01823 pb
Nb of events : 10000

• •Wait: what is l+/l-?

> display multi particles
Multiparticle
all = g u c d
l- = e- mu-
labels:
s u~ c~ d~ s~ a ve vm vt e- mu- ve~ vm~ vt~ e+ mu+ t

•
j = g u c d s u~ c~ d~ s~
Check the run_card… vl = ve vm vt
l+ = e+ mu+
p = g u c d s u~ c~ d~ s~
vl~ = ve~ vm~ vt~

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• Generate the code for dileptonic top decay and compute the
cross-section. Compare with what computed in Ex.1
• > generate p p > t t~, (t > w+ b, w+ > l+ vl), (t~ > w- b~, w- > l- vl~)
• > output myttbardecayed
• > launch
• > set ebeam1 4000
• > set ebeam2 4000
• > set MT 172

• What do we expect? 4/81 = 7.9 Cross-section : 6.936 +- 0.01553 pb

Nb of events : 10000

• Something like 160 * 1/9 = 18 pb? last 1pb of discrepancy comes from
more subtle things

• Wait: what is l+/l-?
• > display multi particles
Multiparticle
all = g u c d
l- = e- mu-
labels: (scales, widths, …)
s u~ c~ d~ s~ a ve vm vt e- mu- ve~ vm~ vt~ e+ mu+ t

• Check the run_card…

j = g u c d s u~ c~ d~ s~
vl = ve vm vt
l+ = e+ mu+
p = g u c d s u~ c~ d~ s~
vl~ = ve~ vm~ vt~

Marco Zaro, 12/14-05-2015 40

 Exercise 2:

Solution
• Questions:
• What is the difference with p p > l+ l- vl vl~ b b~?

Marco Zaro, 12/14-05-2015 41

 Exercise 2:

Solution
• Questions:
• What is the difference with p p > l+ l- vl vl~ b b~?
• It is a much more complex process (will not run in 10s on a laptop)
• Each subprocess has O(100) diagrams rather than O(1)

Marco Zaro, 12/14-05-2015 41

 6
6 ve~
e- 4

Exercise 2:

b~
4 w-
7
b b
5 t 5
vm

Solution
g w+
mu+ page 9/9
mu+
3 1 3

diagram 81 QCD=2, QED=4

• Questions:
2 2
g
g b~ 8 g
b~ 8 5 b~ 8
vm

• What is the difference with p

6 mu- 6
4 w- ve~

e- b~
e-
6 e- p > l+ l- vl vl~ b b~? ve~
6

• It is a much more complex process (will not run in 10s on a laptop)

b~ mu+
ve~ b~ 7
a e+
54 z 3 vm 5 4 e- 4
w+ b
b

• Each subprocess has O(100) diagrams rather than O(1)

3 w-
mu+ b b mu+ b
b a vm 5
3 g 7 3 w+
g 7 g

• This process ‘contains’ t t ̄ decayed, but also other things

mu+ g g > mu+ e- vm ve~ b b~ WEIGHTED=10
1
1 1 3
diagram 84
diagram 77
QCD=2, QED=4
QCD=2, QED=4 diagram 78 QCD=2, QED=4

22 2
7
g g b g 2
6 b~ 8 5 b~ 8 g 7
t b
vm 6
w+ 6 ve~
6 mu+ 6 ve~
4 3 e- 4 b
ve~ w- e-
4
b~ t 7 b~
a
4 e- 4 w-
b 8 w+ vm
b~ 7 5
w-
b t~ b b b mu+
55 z w- ve~
vm 5 6 t 5
w+ vm
g g g w+
b~ 3
mu+
u+ mu+
e- mu+ g 8
33 11 3 1 3
4 1

diagram 8087
diagram QCD=2, QED=4
QCD=2, QED=4 diagram 81 QCD=2, QED=4
diagram 46 QCD=2, QED=4
2 2
2 7
g b
g g
b~ 8 b~ 8
5
5 vm
vm b
5 6 6 w+
ve~ mu- mu+
Marco Zaro, 12/14-05-2015 vm
41 w- ve~ 3
b~ b~ t
vm~ w- e- mu+ e-
 Exercise 2:

Solution
• Questions:
• What is the difference with p p > l+ l- vl vl~ b b~?
• It is a much more complex process (will not run in 10s on a laptop)
• Each subprocess has O(100) diagrams rather than O(1)
• This process ‘contains’ t t ̄ decayed, but also other things
• Which one is correct?

Marco Zaro, 12/14-05-2015 41

 Exercise 2:

Solution
• Questions:
• What is the difference with p p > l+ l- vl vl~ b b~?
• It is a much more complex process (will not run in 10s on a laptop)
• Each subprocess has O(100) diagrams rather than O(1)
• This process ‘contains’ t t ̄ decayed, but also other things
• Which one is correct?
• Strictly speaking t t ̄ decayed, is correct only in the limit Γt=0 i.e. when
tops are on-shell

Marco Zaro, 12/14-05-2015 41

 Exercise 2:

Solution
• Questions:
• What is the difference with p p > l+ l- vl vl~ b b~?
• It is a much more complex process (will not run in 10s on a laptop)
• Each subprocess has O(100) diagrams rather than O(1)
• This process ‘contains’ t t ̄ decayed, but also other things
• Which one is correct?
• Strictly speaking t t ̄ decayed, is correct only in the limit Γt=0 i.e. when
tops are on-shell
• If one searches for (on-shell) top-pair production (e.g.imposing cuts on l,
v, b mass), the full process will give little extra contribution

Marco Zaro, 12/14-05-2015 41

 Exercise 2:

Solution
• Questions:
• What is the difference with p p > l+ l- vl vl~ b b~?
• It is a much more complex process (will not run in 10s on a laptop)
• Each subprocess has O(100) diagrams rather than O(1)
• This process ‘contains’ t t ̄ decayed, but also other things
• Which one is correct?
• Strictly speaking t t ̄ decayed, is correct only in the limit Γt=0 i.e. when
tops are on-shell
• If one searches for (on-shell) top-pair production (e.g.imposing cuts on l,
v, b mass), the full process will give little extra contribution
• If one wants to look away from the resonant region, then the full process
must be used

Marco Zaro, 12/14-05-2015 41

 Exercise 2:

Solution
by terms of O( t /mt ) for inclusive observables. Indeed, similar small-sized di↵er-
ences are observed for di↵erential observables either inclusive in, or insensitive to,
the invariant mass of the (W + , Jb )-system. As an illustrative example we present

• Questions:

in Figure 4 the transverse momentum distribution of the light jet, pT (Jlight ). The
lower panel reveals that the NWA and ET NLO results di↵er by 1-2% in all bins

• What is the difference with p

from the o↵-shell NLO results. In the upper panel it can be seen that both the
p > l+ l- vl vl~ b b~?

NWA and ET results are actually contained within the scale variation band of the

•
NLO o↵-shell result, indicating that for this observable the size of o↵-shell e↵ects
Have a look at single-top production (Papanastasiou et al. arXiv:1305.7088)
is smaller than the scale uncertainty.

Marco Zaro, 12/14-05-2015 42 the reconstructed top quark, M (W + , Jb ).

Figure 5: Invariant mass distribution for
 Part 2: 

NLO

Marco Zaro, 12/14-05-2015 43

 Why should I care?

• Reliable predictions of rates and shapes

• Reliable estimate of uncertainties (scale & PDF)

• Better theoretical accuracy, less need of fine tuning

• Realistic description of the final state

• Better understanding of data

• Steep increase in complexity (in particular for higher
multiplicities)

Marco Zaro, 12/14-05-2015 44

 Why should I care?

• Reliable predictions of rates and shapes

• Reliable estimate of uncertainties (scale & PDF)

• Better theoretical accuracy, less need of fine tuning

• Realistic description of the final state

• Better understanding of data

• Steep increase in complexity (in particular for higher
multiplicities)

Ask a computer to do the hard job

Automation!
Marco Zaro, 12/14-05-2015 44
 NLO: How to?

Marco Zaro, 12/14-05-2015 45

 NLO: How to?

Z
n n n n+1
d N LO = d LO + d V + d 1 d R

Marco Zaro, 12/14-05-2015 45

 NLO: How to?

Z
n n n n+1
d N LO = d LO + d V + d 1 d R

• Virtual and real emission not finite if taken alone

• Infra-red divergences
↵
✓ occur
◆
2 3 19 s
! q q̄g = LO H(✏)CF + +
R
2⇡ ✓ ✏ 2 ◆ ✏2
!
q q̄(g) ↵s 2 3
! V = LO H(✏)CF 2
8
2⇡ ✏ ✏
• Need to include both
✓
in order to
◆
have a finite result
↵s 3
N LO = LO 1 + CF
2⇡ 2
Marco Zaro, 12/14-05-2015 45
 Challenges at NLO:

• Compute (renormalized) one-loop diagrams for any

choice of external particles

• Subtract singularities before doing the integration
(numerically) in d=4

• If showering events, avoid double counting radiation
from the shower and from real emissions

Marco Zaro, 12/14-05-2015 46

 ud→
n=

Marco Zaro, 12/14-05-2015

year

̄ W+ ng
#virt diag
2
0
19
Al
ta 78
re
ll
• NLO evolution:

i, E
llis

1
43
19 ,M
• e.g. pp→W+n jets
ar
Ar
no 89 t in
ld ell
,E i
llis

2
,R

416
20 en
C 0 o

47
am 2
pb
el l
3
20 , Ellis
Bl 0
El ack 9
lis H
, M at
eln +S
4

20 ikov,herpa
4489 57026

Bl 1 Za 

ac 0 nd
kH er
at igh
+S i
5
…

20 herp
Bl 13 a
ac
k H
at
+
A revolution has just happened

Sh
er
pa
 ud→
n=

Marco Zaro, 12/14-05-2015

year

̄ W+ ng
#virt diag
2
0
19
Al
ta 78
re
ll
• NLO evolution:

i, E
llis

1
43
19 ,M
• e.g. pp→W+n jets
ar
Ar
no 89 t in
ld ell
,E i
llis

2
,R

416
20 en
C 0 o

47
am 2
pb
el l
3
20 , Ellis
Bl 0
El ack 9
lis H
, M at
eln +S
4

20 ikov,herpa
4489 57026

Bl 1 Za 

ac 0 nd
kH er
at igh
+S i
5
…

20 herp
Bl 13 a
ac
k H
at
+
A revolution has just happened

Sh
NLO revolution!

er
pa
 Process Syntax Cross section (pb) Process Syntax Cross section (pb)
Vector boson +jets LO 13 TeV NLO 13 TeV Four vector bosons LO 13 TeV NLO 13 TeV
± 5 +15.4% +2.0% 5 +5.2% +1.9%
a.1 pp → W p p > wpm 1.375 ± 0.002 · 10 −16.6% −1.6% 1.773 ± 0.007 · 10 −9.4% −1.6% c.21∗ pp → W + W − W + W − (4f) p p > w+ w- w+ w- 5.721 ± 0.014 · 10−4 +3.7% +2.3%
9.959 ± 0.035 · 10−4 +7.4% +1.7%
−3.5% −1.7% −6.0% −1.2%
+19.7% +1.4% +5.9% +1.3%
a.2 pp → W ± j p p > wpm j 2.045 ± 0.001 · 104 −17.2% −1.1% 2.843 ± 0.010 · 104 −8.0% −1.1% c.22∗ pp → W + W − W ± Z (4f) p p > w+ w- wpm z 6.391 ± 0.076 · 10−4 +4.4% +2.4%
1.188 ± 0.004 · 10−3 +8.4% +1.7%
−4.1% −1.8% −6.8% −1.2%
+24.5% +0.8% +2.4% +0.9%
a.3 pp → W ± jj p p > wpm j j 6.805 ± 0.015 · 103 −18.6% −0.7% 7.786 ± 0.030 · 103 −6.0% −0.8% c.23∗ pp → W + W − W ± γ (4f) p p > w+ w- wpm a 8.115 ± 0.064 · 10−4 +2.5% +2.2%
1.546 ± 0.005 · 10−3 +7.9% +1.5%
+41.0% +0.5% +0.9% +0.6% −2.5% −1.7% −6.3% −1.1%
a.4 pp → W ± jjj p p > wpm j j j 1.821 ± 0.002 · 103 2.005 ± 0.008 · 103 +4.4% +2.4% +7.0% +1.8%
−27.1% −0.5% −6.7% −0.5% c.24∗ pp → W + W − ZZ (4f) p p > w+ w- z z 4.320 ± 0.013 · 10−4 −4.1% −1.7% 7.107 ± 0.020 · 10−4 −5.7% −1.3%
4 +14.6% +2.0% 4 +4.6% +1.9%
a.5 pp → Z p p > z 4.248 ± 0.005 · 10 −15.8% −1.6% 5.410 ± 0.022 · 10 −8.6% −1.5% c.25 ∗ + −
pp → W W Zγ (4f) p p > w+ w- z a 8.403 ± 0.016 · 10 −4 +3.0%
−2.9%
+2.3%
−1.7% 1.483 ± 0.004 · 10−3 +7.2% +1.6%
−5.8% −1.2%
+19.3% +1.2% +5.8% +1.2%
a.6 pp → Zj p p > z j 7.209 ± 0.005 · 103 −17.0% −1.0% 9.742 ± 0.035 · 103 −7.8% −1.0% c.26∗ pp → W + W − γγ (4f) p p > w+ w- a a 5.198 ± 0.012 · 10−4 +0.6% +2.1%
9.381 ± 0.032 · 10−4 +6.7% +1.4%
−0.9% −1.6% −5.3% −1.1%
+24.3% +0.6% +2.5% +0.7%
a.7 pp → Zjj p p > z j j 2.348 ± 0.006 · 103 −18.5% −0.6% 2.665 ± 0.010 · 103 −6.0% −0.7% c.27 ∗
pp → W ZZZ±
p p > wpm z z z 5.862 ± 0.010 · 10 −5 +5.1% +2.4%
1.240 ± 0.004 · 10−4 +9.9% +1.7%
−4.7% −1.8% −8.0% −1.2%
+40.8% +0.5% +1.1% +0.5%
a.8 pp → Zjjj p p > z j j j 6.314 ± 0.008 · 102 6.996 ± 0.028 · 102 c.28∗ pp → W ± ZZγ 1.148 ± 0.003 · 10−4 +3.6% +2.2%
2.945 ± 0.008 · 10−4 +10.8% +1.3%

– 13 –
−27.0% −0.5% −6.8% −0.5% p p > wpm z z a −3.5% −1.7% −8.7% −1.0%
+31.2% +1.7% +24.5% +1.4%
a.9 pp → γj p p > a j 1.964 ± 0.001 · 104 −26.0% −1.8% 5.218 ± 0.025 · 104 −21.4% −1.6% c.29∗ pp → W ± Zγγ p p > wpm z a a 1.054 ± 0.004 · 10−4 +1.7%
−1.9%
+2.1%
−1.7% 3.033 ± 0.010 · 10−4 +10.6% +1.1%
−8.6% −0.8%
+32.8% +0.9% +5.9% +0.8%
a.10 pp → γjj p p > a j j 7.815 ± 0.008 · 103 −24.2% −1.2% 1.004 ± 0.004 · 104 −10.9% −1.2% c.30 ∗
pp → W γγγ±
p p > wpm a a a 3.600 ± 0.013 · 10 −5 +0.4%
−1.0%
+2.0%
−1.6% 1.246 ± 0.005 · 10−4 +9.8% +0.9%
−8.1% −0.8%
+3.8% +2.2% +3.5% +2.2%
Process Syntax Cross section (pb) c.31∗ pp → ZZZZ p p > z z z z 1.989 ± 0.002 · 10−5 −3.6% −1.7% 2.629 ± 0.008 · 10−5 −3.0% −1.7%
+1.9% +2.1% +3.3% +2.1%
c.32∗ pp → ZZZγ p p > z z z a 3.945 ± 0.007 · 10−5 5.224 ± 0.016 · 10−5
Sample of LO and NLO rates for vector-boson production,LO
Vector-boson pair +jets 13 TeV
possibly NLO 13 TeV
within cuts and in association with jets, at the 13-TeV −2.1%
+0.0%
−1.6%
+2.1%
−2.7% −1.6%
+3.4% +2.0%
∗
+ − + 1 +5.0% +2.0%
− 2 +4.0% +1.9% c.33+ pp → ZZγγ p p > z z a a 5.513 ± 0.017 · 10−5 7.518 ± 0.032 · 10−5
hereb.1 pp →the
relevant, W notation
W (4f) understands
p p > w+ the
w- sum of 7.355
the W± 0.005
and·W 10 cross sections,
−6.1% −1.5% and
1.028
wpm± is a
0.003 ·
label
10 that includes
−4.5% −1.4% both W and −0.3% −1.6% −2.6% −1.5%
+2.3% +2.0% +3.4% +1.6%
pp →shell ± sections
0.002 · 10are1 +4.5% +1.9% 1 +3.1% +1.8%
± 0.005 · 10scheme. c.34∗ pp → Zγγγ p p > z a a a 4.790 ± 0.012 · 10−5 7.103 ± 0.026 · 10−5
nedb.2 from the ZZ with define > z= zw+ w-. All1.097
p pwpm cross calculated
−5.6% 1.415
−1.5% in the five-flavour −3.7% Results
−1.4% at the NLO −3.1% −1.6% −3.2% −1.5%
∗ −5 +4.7% +1.9% +7.0% +1.3%
forb.3
W/Z pp → ZW
plus
±
p p > zinwpm
jets are also available MCFM for up±to0.003
2.777 two· jets
+3.6% +2.0%
101 [207–209], ± 0.013 · 101 +4.4%
4.487 heavy-flavour +1.7% c.35 pp → γγγγ p p > a a a a 1.594 ± 0.004 · 10 3.389 ± 0.012 · 10−5
−4.7% −1.5%including identification
−4.4% −1.3% [210–214], −5.7% −1.7% −6.7% −1.3%
1 +22.1% +2.4% 1 +17.6% +2.0%
OWHEGb.4 pp → γγ
[215–217]. NLO cross p psections
> a a for W plus2.510 ± 0.002
three jets· 10
have−22.4%
appeared 6.593
−2.1%in refs. ± 0.021
[218, 219].· 10The−18.8%
BlackHat+SHERPA
−1.9% Process Syntax Cross section (pb)
1 +9.9% +2.0%
b.5 pp → γZ p p > a z 2.523 ± 0.004 · 10 −11.2% −1.6% 3.695 ± 0.013 · 101 +5.4% +1.8%
includes W + and W
tion has provided samples and results for up to Z plus four jets and W plus five jets at the NLO [220–224]. NLO+PS
−7.1% −1.4% 4:merged
Table Heavy quarks
Sample and jets
of NLO rates for four-boson production, possibly withinLO 13at
cuts, TeV NLOthat
the LHC 13 TeV. wpm is a label 13 TeV
b.6 pp → γW ± p p > a wpm 2.954 ± 0.005 · 101 +9.5% +2.0%
7.124 ± 0.026 · 10 1 +9.7% +1.5%
andpresented
is defined via wpm = w+ w-. All cross sections calculated in the 5-flavor scheme,+0.8%
except the processes with at least two W -bosons
or W plus up to three jets are also available in SHERPA [225]. γ −11.0% plus up to three jets calculations−9.9%
−1.7% have−1.3%
been d.1 in→define
pp jj p p > j j 1.162 ± 0.001 · 106 +24.9% 1.580 ± 0.007 · 106 +8.4% +0.7%
+ − 1 +11.6% +1.0% 1 ±+4.9% +1.1%prevent top resonant contributions from appearing at NLO. For all processes in this table NLO
−18.8% QCD corrections have never been
−0.9% computed
−9.0% −0.9% befor
b.7 Weppdo
, 227]. →Wnot Wshowj (4f)
cross psections
p > w+ for 2.865 ± V0.003
w- EW-induced
j plus· 10two −10.0%
jets processes
−0.8% 3.730 ±
with = γ, ·Z,
V 0.013 10W −4.9%
, which are available
−0.8% d.2 in→ jjj
pp p p > j j j 8.940 ± 0.021 · 104 +43.8% +1.2%
7.791 ± 0.037 · 104 +2.1% +1.1%
−28.4% −1.4% −23.2% −1.3%
0 +10.9% +1.0% 0 +5.0% +1.1%
b.8 and
O [228] pp → ZZjbeen studiedp in
have p ref.
> z [229].
z j 3.662 ± 0.003 · 10 −9.3% −0.8% 4.830 ± 0.016 · 10 −4.8% −0.9%
3 +25.2% +1.5% 3 +15.9% +1.5%
+11.6% +0.9% +4.9% +0.9% d.3 pp → bb̄ (4f) p p > b b∼ 3.743 ± 0.004 · 10 6.438 ± 0.028 · 10
b.9 pp → ZW ± j p p > z wpm j 1.605 ± 0.005 · 101 −10.0% −0.7% 2.086 ± 0.007 · 101 −4.8% −0.7%
−18.9% −1.8%
+44.1% +1.6%
−13.3% −1.7%
+6.8% +1.5%
+20.3% +1.2% +17.2% +1.0% d.4∗ pp → bb̄j (4f) p p > b b∼ j 1.050 ± 0.002 · 103 1.327 ± 0.007 · 103
b.10 pp → γγj p p > a a j 1.022 ± 0.001 · 101 −17.7% −1.5% 2.292 ± 0.010 · 101 −15.1% −1.4%
−28.5% −1.8%
+61.8% +2.1%
−11.6% −1.8%
+8.2% +2.0%
+14.5% +1.0% +7.3% +0.9% d.5∗ pp → bb̄jj (4f) p p > b b∼ j j 1.852 ± 0.006 · 102 2.471 ± 0.012 · 102
b.11∗ pp → γZj p p > a z j 8.310 ± 0.017 · 100 −12.8% −1.0% 1.220 ± 0.005 · 101 −7.4% −0.9%
−35.6% −2.4%
+61.7% +2.9%
−16.4% −2.3%
+20.9% +2.9%
+13.7% +0.9% +7.2% +0.9% d.6 pp → bb̄bb̄ (4f) p p > b b∼ b b∼ 5.050 ± 0.007 · 10−1 8.736 ± 0.034 · 10−1
b.12∗ pp → γW ± j p p > a wpm j 2.546 ± 0.010 · 101 −12.1% −1.0% 3.713 ± 0.015 · 101 −7.1% −1.0%
−35.6% −3.4% −22.0% −3.4%

+25.4% +2.1% +10.5% +2.2% d.7 pp → tt̄ p p > t t∼ 4.584 ± 0.003 · 102 +29.0% +1.8%
6.741 ± 0.023 · 102 +9.8% +1.8%
b.13 pp → W + W + jj p p > w+ w+ j j 1.484 ± 0.006 · 10−1 −18.9% −1.5% 2.251 ± 0.011 · 10−1
−10.6% −1.6%
−21.1% −2.0%
+45.1% +2.2%
−10.9% −2.1%
+25.4% +2.4% +10.1% +2.5% d.8 pp → tt̄j p p > t t∼ j 3.135 ± 0.002 · 102 −29.0% −2.5% 4.106 ± 0.015 · 102 −12.2% +2.1%
+8.1%
b.14 pp → W − W − jj p p > w- w- j j 6.752 ± 0.007 · 10−2 −18.9% −1.7% 1.003 ± 0.003 · 10−1
−10.4% −1.8%
−2.5%
+27.2% +0.7% +5.0% +0.7% d.9 pp → tt̄jj p p > t t∼ j j 1.361 ± 0.001 · 102 +61.4% +2.6%
1.795 ± 0.006 · 102 +9.3% +2.4%
b.15 pp → W + W − jj (4f) p p > w+ w- j j 1.144 ± 0.002 · 101 −19.9% −0.5% 1.396 ± 0.005 · 101
−6.8% −0.6%
−35.6% −3.0%
−3 +63.8% +5.4%
−16.1% −2.9%
−3 +30.8% +5.5%
+26.6% +0.7% – 78 –
+5.8% +0.8% d.10 pp → tt̄tt̄ p p > t t∼ t t∼ 4.505 ± 0.005 · 10 9.201 ± 0.028 · 10
b.16 pp → ZZjj p p > z z j j 1.344 ± 0.002 · 100 −19.6% −0.6% 1.706 ± 0.011 · 100
−7.2% −0.6%
−36.5% −5.7% −25.6% −5.9%
+26.7% +0.7% +3.1% +0.7% 0 +62.1% +2.9% 1 +37.6% +2.9%
b.17 pp → ZW ± jj p p > z wpm j j 8.038 ± 0.009 · 100 −19.7% −0.5% 9.139 ± 0.031 · 100
−5.1% −0.5%
d.11 pp → t t̄b b̄ (4f) p p > t t∼ b b∼ 6.119 ± 0.004 · 10 −35.7% −3.5% 1.452 ± 0.005 · 10 −27.5% −3.5%
+26.2% +0.6% +8.8% +0.6%
b.18 pp → γγjj p p > a a j j 5.377 ± 0.029 · 100 −19.8% −1.0% 7.501 ± 0.032 · 100
−10.1% −1.0% Process Syntax Cross section (pb)
+24.3% +0.6% +6.5% +0.6%
b.19∗ pp → γZjj p p > a z j j 3.260 ± 0.009 · 100 4.242 ± 0.016 · 100
−18.4% −0.6%
+24.7% +0.6%
Table 5: Heavy
−7.3% −0.6%
+3.6% +0.6%
Samplequarks+vector
of LO and NLO bosonstotal rates for the production of heavy LOquarks
13 TeVand/or jets, possibly within NLO cuts,
13 TeVat the 13-TeV L
b.20∗ pp → γW ± jj p p > a wpm j j 1.233 ± 0.002 · 101 −18.6% −0.6% 1.448 ± 0.005 · 101
−5.4% −0.7% 2 +42.3% +2.0% 2 +29.8% +1.5%
we also report
e.1 the
pp →integration
W bb̄ (4f) errors, and
±
p p the
> wpmfractional
b b∼ scale (left)
3.074 and ·PDF
± 0.002 10 (right) uncertainties.
−29.2% −1.6% 8.162 ±Processes
0.034 · 10 d.1 and d.2, as we
−23.6% −1.2%
Process Syntax Cross section (pb) processese.2
involving
pp →atZ least
bb̄ (4f)a top pair, pare
p computed
> z b b∼ in the 6.993
five-flavour · 102 +33.5%
± 0.003scheme. +1.0%
Processes
−24.4% −1.4% that explicitly
1.235 ± 0.004 involve
· 10 3 +19.9% +1.0%
b-quarks in the
−17.4% −1.4%
Three vector bosons +jet LO 13 TeV NLO 13 TeV 3 +51.9% +1.6% 3 +33.7% +1.4%
2: Sample of LO and NLO rates for vector-boson pair state are e.3
production, possibly within cuts and in association with jets, at the 13-TeV pp
calculated→ γ
in b b̄
the(4f)
four-flavour pscheme.
p > a bFor
b∼ 1.731
processes ± 0.001
d.3–d.6 we· 10
require 2 (or 4)
−34.8% −2.1% 4.171
b-jets in ± 0.015
the final· 10
state with |η| < 2.5.
−27.1% −1.9%

we c.1
also report
pp → Wthe
+ integration
W − W ± (4f) errors,
p p >and thewpm
w+ w- fractional scale
1.307 (left)· 10
± 0.003 and +0.0% +2.0%
−1 PDF (right) uncertainties.
−0.3% −1.5% 2.109 ± 0.006 See
· 10−1table processes
+5.1% +1.6%
1 for the meaning
−4.1% −1.2%
d.1–d.6,
e.4 ∗ weWrequire
pp → ± the (b)-jets
bb̄ j (4f) p pto> have
wpm bpTb∼> j80 GeV,
1.861 with at ·least
± 0.003 102 one of +0.7%
+42.5% them with pT ±>0.013
3.957 100 ·GeV. Calculations
102 +27.0% +0.7% of c
−27.7% −0.7% −21.0% −0.6%
All → ZW + W
c.2crossppsections are− calculated
(4f) p in
p >the five-flavour
z w+ w- scheme, except
9.658 ± 0.065 −2
· 10for +0.8% +2.1%
processes
−1.1% −1.6%b.1, b.7, and
1.679 ± b.15,
0.005 · which
10 are sections
−1 +6.3% +1.6%
obtained
−5.1% −1.2%
at
in
e.5the
∗ NLO for this
pp → Z bb̄ j (4f) class of processes are
p p > z b b∼ j available in the literature 2 as well
+42.4%
1.604 ± 0.001 · 10 −27.6% −1.1% as in
+0.9% public codes: from 2 the seminal
+21.0%
2.805 ± 0.009 · 10 −17.6% −1.0% +0.8%result
+1.0% +2.0% −2 +6.8% +1.5%
avour pp →to
c.3 scheme ZZW ±
avoid resonant-top
p p contributions.
> z z wpm 2.996
NLO ± 0.016
results V−2
for· 10 V production
−1.4% −1.6% have 5.550 ± 0.020
been known · 10for some the hadroproduction
−5.5% time
−1.1% [243–252],e.6∗ pp → γofbb̄aj heavy
(4f) quark p pair
p >[282–286],
a b b∼ jto their NLO+PS
7.812 ± 0.017 ·implementation
102 +51.2% +1.0%in MC@NLO [169] and
−32.0% −1.5%
POWHEG
1.233 ± 0.004 · 103 +18.9% +1.0% [287
−19.9% −1.5%
−2 +0.0% +1.9% −2 +2.7% +1.9%
pp → ZZZ
c.4 available
blicly in MCFM and inp VBFNLO
p > z z z [241], and are±matched
1.085 0.002 · 10to parton showers 1.417 ± 0.005 · 10
in MC@NLO [24] and tt̄j [288] (also
POWHEG [253].including top decays [254,289] and parton shower effects [290]), −1 to the computation
+23.9% +2.1% of tt̄jj [291]. Merged NLO+PS re
−0.5% −1.5%
−1 +1.9% +2.0%
−2.1% −1.5%
e.7 pp → tt̄ W ± p p > t t∼ wpm 3.777 ± 0.003 · 10 5.662 ± 0.021 · 10−1 +11.2% +1.7%

esults V→
c.5 for pp V γW
+ −
withWup (4f) p p jet
to an extra > ahave
w+ w-been made1.427 ± 0.011 in
available · 10POWHEG 2.581 ±
[254, 255].
−2.6% −1.5% NLO · 10−1 +5.4%
0.008corrections +1.4%
for
γγ tplus
to−1.1%
−4.3% t̄ plusupjets to are also available [188, 292, 293]. NLO results for three jets [294], −18.0% −1.6%
four jets
−1 +30.5% +1.8%[73], and up to five jets −10.6% −1.3%
−1 [295,
+9.7%296]
+1.9%have
± −2 +4.4% +1.9% +7.6% +1.0% e.8 pp → tt̄ Z p p > t t∼ z 5.273 ± 0.004 · 10 −21.8% −2.1% 7.598 ± 0.026 · 10 −11.1% −2.2%
ets c.6
are also → γγW [256–260].
ppknown p p available
Other 2.681
> a a wpmresults are: W±±0.007
W ± jj· 10[261],−5.6%
W ±W ± jj (EW+QCD) · 10−2Zγj
8.251 ± 0.032[262], published.
[263], W γjj e.9 Two-ppand
[264], three-jet event generation is available in POWHEG [297, 298].+1.6%
0 +29.6% Calculations for bb̄bb̄ [299, 300], +1.7%
tt̄bb̄ [301–3
±
−1.6%
−2 +0.8% +1.9%
−7.0% −1.0%
−1 +7.2% +1.2%
→ t t̄ γ p p > t t∼ a 1.204 ± 0.001 · 10 −21.3% −1.8% 1.744 ± 0.005 · 100 +9.8%−11.0% −2.0%
c.7 pp → γZW p p > a z wpm 4.994 ± 0.011 · 10 1.117 ± 0.004 · 10 and
[265], W γj [266, 267], W Zj [268]. We do not show results for NLO corrections to EW-induced production −5.9%
−1.4% −1.6% plustt̄ttwo
of V V−0.9% t̄ [304]
jets,production have appeared in the literature. +40.9% +1.3% +11.2% +1.2%
c.8 pp → γZZ p p > a z z 2.320 ± 0.005 · 10−2 +2.0% +1.9%
3.118 ± 0.012 · 10−2 +2.8% +1.8% e.10∗ pp → tt̄ W ± j p p > t t∼ wpm j 2.352 ± 0.002 · 10−1 3.404 ± 0.011 · 10−1
W ± W ∓ jj [269], W Zjj [270], and ZZjj [271], which can also be obtained with
−2.9% POWHEG and VBFNLO.
−1.5% −2.7% −1.4% −27.1% −1.0% −14.0% −0.9%
– 79 –

+46.2% +2.7% +7.0% +2.5%

c.9 pp → γγZ p p > a a z −2 +5.6% +1.9%
3.078 ± 0.007 · 10−2 +4.5%
−6.8% −1.6% 4.634 ± 0.020 · 10 +1.7%
−5.0% −1.3%
e.11∗ pp → tt̄ Zj p p > t t∼ z j 3.953 ± 0.004 · 10−1 −29.5% −3.0% 5.074 ± 0.016 · 10−1 −12.3% −2.9%
+45.4% +2.3% +7.5% +2.2%
c.10 pp → γγγ p p > a a a 1.269 ± 0.003 · 10−2 +9.8% +2.0%
−11.0% −1.8% 3.441 ± 0.012 · 10−2 +11.8% +1.4%
−11.6% −1.5%
e.12∗ pp → tt̄ γj p p > t t∼ a j 8.726 ± 0.010 · 10−1 −29.1% −2.6% 1.135 ± 0.004 · 100 −12.2% −2.5%
+30.9% +2.1% +10.9% +2.1%
c.11 pp → W + W − W ± j (4f) p p > w+ w- wpm j 9.167 ± 0.010 · 10−2 +15.0%
−12.2%
+1.0%
−0.7% 1.197 ± 0.004 · 10−1 +5.2% +1.0%
−5.6% −0.8%
e.13∗ pp → tt̄ W − W + (4f) p p > t t∼ w+ w- 6.675 ± 0.006 · 10−3 −21.9% −2.0% 9.904 ± 0.026 · 10−3 −11.8% −2.1%
+26.6% +2.5% +10.6% +2.3%
c.12∗ pp → ZW + W − j (4f) p p > z w+ w- j 8.340 ± 0.010 · 10−2 +15.6%
−12.6%
+1.0%
−0.7% 1.066 ± 0.003 · 10−1 +4.5% +1.0%
−5.3% −0.7%
e.14∗ pp → tt̄ W ± Z p p > t t∼ wpm z 2.404 ± 0.002 · 10−3 −19.6% −1.8% 3.525 ± 0.010 · 10−3 −10.8% −1.6%
+25.4% +2.3% +10.3% +2.0%
c.13∗ pp → ZZW ± j p p > z z wpm j 2.810 ± 0.004 · 10−2 +16.1%
−13.0%
+1.0%
−0.7% 3.660 ± 0.013 · 10−2 +4.8% +1.0%
−5.6% −0.7%
e.15∗ pp → tt̄ W ± γ p p > t t∼ wpm a 2.718 ± 0.003 · 10−3 −18.9% −1.8% 3.927 ± 0.013 · 10−3 −10.4% −1.5%
+29.3% +1.7% +7.9% +1.7%
c.14∗ pp → ZZZj p p > z z z j 4.823 ± 0.011 · 10−3 +14.3%
−11.8%
+1.4%
−1.0% 6.341 ± 0.025 · 10−3 +4.9% +1.4%
−5.4% −1.0%
e.16∗ pp → tt̄ ZZ p p > t t∼ z z 1.349 ± 0.014 · 10−3 −21.1% −1.5% 1.840 ± 0.007 · 10−3 −9.9% −1.5%
+30.1% +1.7% +9.7% +1.8%
c.15∗ pp → γW + W − j (4f) p p > a w+ w- j 1.182 ± 0.004 · 10−1 +13.4%
−11.2%
+0.8%
−0.7% 1.233 ± 0.004 · 103 +18.9% +1.0%
−19.9% −1.5%
e.17∗ pp → tt̄ Zγ p p > t t∼ z a 2.548 ± 0.003 · 10−3 −21.5% −1.6% 3.656 ± 0.012 · 10−3 −11.0% −1.9%
+28.4% +1.3% +7.8% +1.4%
c.16 pp → γγW ± j p p > a a wpm j 4.107 ± 0.015 · 10−2 +11.8%
−10.2%
+0.6%
−0.8% 5.807 ± 0.023 · 10−2 +5.8% +0.7%
−5.5% −0.7%
e.18∗ pp → tt̄ γγ p p > t t∼ a a 3.272 ± 0.006 · 10−3 −20.6% −1.1% 4.402 ± 0.015 · 10−3 −9.7% −1.4%
+14.4% +0.7% +5.1% +0.8%
c.17∗ pp → γZW ± j p p > a z wpm j 5.833 ± 0.023 · 10−2 7.764 ± 0.025 · 10−2
Marco Zaro, 12/14-05-2015 486:
−12.0% −0.6% −5.5% −0.6% Process Syntax Cross section (pb)
+12.5% +1.2% +5.6% +1.2%
c.18∗ pp → γZZj p p > a z z j 9.995 ± 0.013 · 10−3 −10.6% −0.9% 1.371 ± 0.005 · 10−2 −5.5% −0.9% Single-top LO 13 TeV NLO 13 TeV
+10.9% +1.0% Table
+7.0% +1.0% Sample of LO and NLO total rates for the production of heavy quarks in association with vector bosons, possibly wit
c.19∗ pp → γγZj p p > a a z j 1.372 ± 0.003 · 10−2 −9.4% −0.9% 2.051 ± 0.011 · 10−2 −6.3% −0.9% 2 +9.4% +0.4% +1.4% +0.4%
f.1 pp → tj (t-channel) p p > tt j $$ w+ w- 1.520 ± 0.001 · 10 1.563 ± 0.005 · 102
 Loop proliferation

• in the last 10 years, many different techniques have been

developed in order to compute any one-loop process.

• They (roughly) fall into 3 classes

• Tensor reduction

• Generalized unitarity
Passarino,Veltman,1979

Denner, Dittmaier, hep-ph/509141

• Integrand reduction Binoth, Guillet, Heinrich, Pilon, Reiter, arXiv:0810.0992

Bern, Dixon, Dunbar, Kosower, hep-ph/9403226 + …

Ellis, Giele, Kunszt, arXiv:0708.2398

+ Melnikov, arXiv:0806.3467

Ossola, Papadopoulos, Pittau, hep-ph/0609007

Del Aguila, Pittau, hep-ph/0404120

Mastrolia, Ossola, Reiter, Tramontano, arXiv:1006.0710

Marco Zaro, 12/14-05-2015 49

 Basics of loops:
2
|M | = T r[p/2 γ p/1 γ ] µ g4
ν
T r[(p/3 + m)γ µ (p/4 − m)γ ν ] =

Passarino-Veltman reduction
(p1 · p2 ) 2

s1 s2 s3 s4 s2 g µν
s1
|M | 2
= g v̄(p2 ) γµ u(p1 ) ū(p3 )s3 γν u(p4 )s
p1 · p2
! 2
|M |2 = (|M |s1 s2 s3 s4 )

• Any one-loop amplitude can be written as a linear combination s1 s2 s3 s4

∼ αs2 ∼ αe2

of a basis of integrals
αe
αs
! 0.1
m
X1 " D
d q 1
! A(q) =
i0 <i1 <i2 <i3
d(i0 i1 i2 i3 )D0 (i0 i1 i2 i3 )
k1
q + k1
k2 D(i1 , i2 , i3 , i4 ) =
N Di1 Di2 Di3 Di4
.. k3 k4 " D
m
X1 q D1 . d q 1
! + c(i0 i1 i2 )C0 (i0 i1 i2 ) kn D0 D2 k5 C(i1 , i2 , i3 ) =
N Di1 Di2 Di3

q + . . . + k5
i0 <i1 <i2
Dm 1 " D
! +
m
X1
b(i0 i1 )B0 (i0 i1 )
D3
B(i1 , i2 ) =
d q 1
N Di1 Di2
i0 <i1 " D
! m
X1 A(i1 ) =
d q 1
+ a(i0 )A0 (i0 ) N Di1
! i0

• Integrals can be computed once for all and coded into libraries

+ R

QCDLoops, Ellis, Zanderighi

! OneLoop,Van Hameren

• Coefficients can be found by computing analytically the …

amplitude and solving (algebraically) a system of equations

• In practice feasible only for low multiplicities
Marco Zaro, 12/14-05-2015 50
 Passarino-Veltman reduction

at the integrand level

Ossola, Papadopoulos, Pittau, hep-ph/0609007

CutTools: Ossola, Papadopoulos, Pittau, arXiv:0711.3596

• The integrand of the loop amplitude can be written as

!
!
a,b,c,d
! are the same as in the
! ̃ ̃ slide; 

previous
ã,b,c̃,d gives 0 after integration
!
(spurious terms)
!
!
• The system of equation can be cast in a triangolar form and
solved numerically at each point in the phase-space:

• Fix external momenta

• Choose q such that all D’s vanish but D1,D2,D3,D4 → get the coefficient d(1234)

• Do that for all 4-point integrals, then for 3-point ones, until all coefficients are
known
Marco Zaro, 12/14-05-2015 51
 The evil is in the details

• Numerical approach: must work in D=4

• But scalar integrals are computed in D=D

• Mismatch gives origin to the rational terms (R) which need to
be added in order to get the correct result

• Two kind of rational terms exist:

• Rational terms that originates from the denominators (R1):

can be obtained from the amplitude without extra infos

• Rational terms that originates from the numerator (R2):

need to be added to the amplitude, can be provided as extra
Feynman rules

• Are in a finite number

• Can be computed once for all from the model

• UV renormalisation done via extra Feynman rules as well
Marco Zaro, 12/14-05-2015 52
 Subtraction of Infra-Red divergences:

Problem #1
• Real and virtual contributions are not separately finite

• Numerical integration (in D=4): can integrate only finite quantities

• QCD helps: divergences have an universal structure:

p • Soft:
X pp 2 i j
|Mnij |2
k

lim |Mn+1 | '

+

! pi k p j k
p

k!0
k
• Collinear:
ij

! lim |Mn+1 |2 ' |Mn |2 P AP (z)

p//k
2
(p + k) = 2Ep Ek (1 cos ✓!pk )
• Use universality of limits to build local counterterms to render n
and n+1 body contributionsZfinite Z
n n n n+1
d N LO =d LO +d V d 1 C+ d 1 C +d R
Integrating C is much simpler than than R

Can be done in D=D dimension (once and for all)
Marco Zaro, 12/14-05-2015 53
 Subtraction of Infra-Red divergences:

Problem #2
• Real emissions can have singularities in different regions of the
phase-space:

• E.g. gg→t tg:

̄
• g collinear to g or g

• g soft

• Numeric integrators (VEGAS) are quite dumb (still, that is the best
one can do): peaks need to be well aligned with the integration
variables

• “Divide et impera” solution: integrate one singularity at the time, with
the most suitable phase-space parameterisation:
2
X 2
X 2
X
|M | = Sij |M | = |M |ij Sij = 1
ij ij
Sij ! 1 if ki · kj ! 0 Sij ! 0 if km6=i · kn6=j ! 0

Marco Zaro, 12/14-05-2015 54

 Advantages:

• Parallelization: Each contribution can be integrated independently,

with a suitable PS parameterization

• The number of contribution grows at most as n2

• Symmetries can be used to reduce the numbers of contributions

• E.g: only 3 contributions for gg→g…g

Marco Zaro, 12/14-05-2015 55

 In other words, ξi is the rescaled energy of the FKS parton, and yij is the cosine of
the angle between the FKS parton and its sister. Using these variables, the soft and

More details…
collinear singularities of Sij M(n+1,0) correspond to ξi = 0 and to yij = 1 respectively
(note, however, that the matrix element may not be singular in either or both of
these limits, depending on the identities of particles i and j). We have (see ref. [12])
! (n+1)
! ij
dσ (n+1)
The actual form for dφn depends on =
(r) the specific parametrization
dσij (r) , adopted for dφn+1(4.28)
,
• Local
butsubtraction is
the following equationsdone as
must always a modified ‘+’ prescription

hold: (i,j)∈PFKS

• Subtract only close enough to the singularity

! = s ij
where lim d φ dφ n if m = 0 ,
3
(4.31)
n i
(4π)
ξi →0
# $ #lim dφ!ij$=% s dφ if m = m &= 0 . n
(B)
(4.32)
! 1 yij →11
n
(4π) 3 n i j J L
dξidyij dϕi dφ'ij
(n+1) 2 (n+1,0)
dσij (r) = (1 − yij )ξi M (r) Sij (r) n .
ξi entering
The distributions c 1 − yeq.
ij (4.29)
δ are defined as follows10 : N (r)
! # $ (4.29)
" ξmax " ξmax
1 f (ξi) − f (0)Θ(ξcut − ξi )
! The variable ϕi parametrizes
dξif (ξi) the azimuthal
= dξi direction of the FKS ,parton, but its
(4.33)
0 ξi c 0 ' ij ξi
precise definition
" 1 is irrelevant
# here.
$ The" 1quantity d φ n is related to the n-body phase
! space, and is implicitly 1
dyij g(yij ) defined as follows:
= dyij
g(yij ) − g(1)Θ(yij − 1 + δ)
, (4.34)
−1 1 − yij δ −1 1 − yij
! 'ij .
• This is (a summary of)
where dφn+1 = ξi dξ%i dyij dϕi'
dφ (4.30)
the so-called FKS-subtraction 

n+3 2n
1 &
9
We remind the reader that multiple − and collinear
ξmax = 1soft mk .
singularities, (4.35) that
i.e. configurations
(Frixione, Kunszt, Signer, hep-ph/9512328) n(B) s
would contribute to NNLO, are cut by J L k=3 .
In eq. (4.33) and (4.34) ξcut and δ are free parameters, that can be chosen in the
ranges
0 < ξcut ≤ ξmax , 0 < δ ≤ 2. (4.36)
It would be possible to choose different 16 values of ξcut and δ for each of the cross
(n+1)
sections dσij contributing to eq. (4.28). In practice, we shall choose the same
Marco Zaro, 12/14-05-2015 56
value of ξ for all (i, j) pairs, the same value of δ for all pairs (i, j) with j ≥ 3 (which
 Why to care about showering?

Marco Zaro, 12/14-05-2015 57

 Why to care about showering?
• Quarks and gluons undergo confinement: need to a description of
final states in terms of hadrons

VS

Marco Zaro, 12/14-05-2015 57

 Why to care about showering?
• Quarks and gluons undergo confinement: need to a description of
final states in terms of hadrons
• Parton shower cures bad behaviours of fixed-order computations
and resums soft logarithms Frixione, Hirschi, Pagani, Shao, Zaro, arXiv:1407.0823 10
-2

tt-H production at the 13 TeV LHC

boosted cuts: pT(t), pT(t-), pT(h) > 200 GeV
104
+ 4 LO
W prod. at the 13 TeV LHC 10 LO+NLO QCD
10-3 LO+NLO QCD+Weak

σ per bin [pb]

10
3 103

MadGraph5_aMC@NLO
10-4
σ per bin [pb]

102
102 MadGraph5_aMC@NLO

101
-5
10
ratio over LO
0 5 10 15 20 1.6
1 fNLO 1.2
10 NLO+HW6
0.8
0.8 relative contributions

0.4
0 QCD Weak HBR
10 0
0 50 100 150 200 -0.4
pT(W) 0 200 400 600 800
-
pT(tt) [GeV]
Marco Zaro, 12/14-05-2015 57
 (Unweighted) event generation

and matching to parton-showers
MC@NLO: Frixione, Webber hep-ph/0204244
• Problem #1: the n and n+1 body cross-sections are not separately
finite at NLO

• Cannot unweight an infinite cross-section

• Problem #2: when showering events, one must not double count
radiation from the shower and the real-emission matrix-element

• Solution: introduce
d
Z
“M C@N LO”
the so-called
Z Monte-Carlo
Z counterterms

n n+1
= d n (B +V + d 1 M C) IM C (O) + d n+1 (R M C) IM C (O)
! dO
S-events H-events
• They are related to the shower Sudakov

 Z
! k MC MC
IM C = + d 1 + ... = exp d 1
B B
! 1 ↵s
MC = J P (z M C )B
! tM C 2⇡
• MC is shower-dependent (i.e. the same sample of event has to be
showered with a specific parton
58
shower)
Marco Zaro, 12/14-05-2015
 aMC@NLO

• Can write the code for computing cross-section at NLO for any
process (limited only by CPU time)

• 2 main run modes (at LO and NLO):

• fixed-order

• do not generate events. Just compute the cross-section and optionally
fill histograms on the fly

• PS matching

• generate events à la MC@NLO. Distributions obtained from NLO
events are unphysical unless events are showered

• In either case, the cross-section has to be the same within
statistical uncertainties

Marco Zaro, 12/14-05-2015 59

 Including the decay in NLO

samples
• How to deal with unstable particles (e.g. top) at the NLO?

• Cannot use decay-chain syntax: gauge invariance violated at NLO

• Very rough solution:

• Let the shower decay the particles: spin correlations are lost

• Very refined solution:

• Generate process with only stable particles (pp>l+l-vv~bb~):
includes spin correlations, off-shell effects, non resonant
contributions, ...

• Needs special treatment of intermediate resonances (e.g. complex-mass scheme)

• Computationally very expensive

• Only needed when background is enhanced or when aiming at very high precision

Marco Zaro, 12/14-05-2015 60

 Including the decay in NLO

samples
• How to deal with unstable particles (e.g. top) at the NLO?

• Cannot use decay-chain syntax: gauge invariance violated at NLO

• Very rough solution:

• Let the shower decay the particles: spin correlations are lost

• Very refined solution:

• Generate process with only stable particles (pp>l+l-vv~bb~):
includes spin correlations, off-shell effects, non resonant
contributions, ...

• Needs special treatment of intermediate resonances (e.g. complex-mass scheme)

• Computationally very expensive

• Only needed when background is enhanced or when aiming at very high precision

Anything in between?
Marco Zaro, 12/14-05-2015 60
 Spin correlations made easy:
MadSpin
Artoisenet, Frederix, Mattelaer, Rietkerk, arXiv:1212.3460

Marco Zaro, 12/14-05-2015 61

 Spin correlations made easy:
MadSpin
Artoisenet, Frederix, Mattelaer, Rietkerk, arXiv:1212.3460

• Wish-list:

• For a given event sample (LO or MC@NLO), include the decay of
any final state particle

• Keep spin correlations

• Generate decayed unweighted events

Marco Zaro, 12/14-05-2015 61

 Spin correlations made easy:
MadSpin
Artoisenet, Frederix, Mattelaer, Rietkerk, arXiv:1212.3460

• Wish-list:

• For a given event sample (LO or MC@NLO), include the decay of
any final state particle

• Keep spin correlations

• Generate decayed unweighted events
• Solution:

• Read event

• Generate decay kinematics

• Reweight the event with ratio
|MP +D | / |MP |
2 2

• Or do secondary unweighting

• Generate many decay configurations
2 2
⇣ until

2 2
⌘
|MP +D | / |MP | > Rand() max |MP+D | / |MP |

Marco Zaro, 12/14-05-2015 61

 Spin correlations made easy:
MadSpin
Artoisenet, Frederix, Mattelaer, Rietkerk, arXiv:1212.3460

• Wish-list:

• For a given event sample (LO or MC@NLO), include the decay of
any final state particle

• Keep spin correlations

• Generate decayed unweighted events
• Solution:

• Read event

• Generate decay kinematics

• Reweight the event with ratio
|MP +D | / |MP |
2 2

• Or do secondary unweighting

• Generate many decay configurations
2 2
⇣ until

2 2
⌘
|MP +D | / |MP | > Rand() max |MP+D | / |MP |
• This was been done for the first time for t t ̄ and singletop
Frixione, Leanen, Motylinski, Webber, arXiv:hep-ph/0702198
Marco Zaro, 12/14-05-2015 61
 Spin correlations made easy:
MadSpin
Artoisenet, Frederix, Mattelaer, Rietkerk, arXiv:1212.3460

• How to deal with MC@NLO events?

• Spin correlations usually have tiny effects on observables

• Include them at tree level

• For H (n+1 body) events, use decayed real-emission matrix-
element

• For S (n body) events, use decaysed born matrix-element

• This guarantees NLO accuracy for observables related to
production (e.g. top pt)

• This includes all spin correlation for observables related to
production + decay (apart non-factorizable ones)

Marco Zaro, 12/14-05-2015 62

. (4.2) features only 3-resonant contributions (see
resonant and non-resonant diagrams), the results
ples are basically the same if one considers only the Spin correlations made easy:
MadSpin
eco cuts together; for this reason, we shall discuss
and, one of the reasons for comparing eqs. (4.1)
he importance of non-3-resonant contributions
Artoisenet, to Rietkerk, arXiv:1212.3460
Frederix, Mattelaer,

lts. •
case we shall present both the generation-cut-only
Example: 6 lepton production at NLO (arXiv:1405.0301)
-6 -5 -6
4.01010 3.0 10
e+ee+-eµ-+µν+µντµ-ντ-τνproduction at the 13 TeV LHC
τ production at the 13 TeV LHC
+ - + -
e e µ νµτ ντ production at the 13 TeV LHC
-6
3.5 10
NLO, ME LO, ME (x1.7)
-6
-6 -6 2.5 10 NLO, ME V-reco LO, ME V-reco (x1.85)
3.01010 NLO, MS V-reco LO, MS V-reco (x1.85)
NLO, ME LO, ME NLO, PSMC V-reco LO, PSMC V-reco (x1.85)
-6
σ per bin [pb]
bin [pb]

σ per bin [pb]

NLO, ME V-reco LO, ME V-reco
2.5 10
NLO, MS V-reco LO, MS V-reco -6
2.0 10
NLO, PSMC V-reco LO, PSMC V-reco
-6
2.01010
-7
σ per
MadGraph5_aMC@NLO

MadGraph5_aMC@NLO
MadGraph5_aMC@NLO

MadGraph5_aMC@NLO
-6 -6
1.5 10 1.5 10
-6
1.0 10 NLO, ME LO, ME (x0.5)
10-8 NLO, ME V-reco LO, ME V-reco (x0.5)
-6
-7 NLO, MS V-reco LO, MS V-reco (x0.5) 1.0 10
5.0 10 NLO, PSMC V-reco LO, PSMC V-reco (x0.5)

Ratio
Ratioover NLO,
over ME
NLO, V-reco
ME V-reco Ratio over NLO, ME V-reco
1.4
1.2 1.4
1.2
1 1.2
1 1
1
0.7 Ratios
RatiosLOLO
over NLO
over (LO
NLO notnot
(LO resc.)
resc.) 0.7 Ratios LO over NLO (LO not resc.)
0.8
0.6
0.6 0.6
0.4
0.2
0.5 0.5
0
3 -1250 -0.8 500
-0.6 750 -0.2
-0.4 1000 0 1250 0.2 1500
0.4 17500.8 2000
0.6 1 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
+ - + - + -
m(e e µ cosψ
νµτ νµτ)τ[GeV] cosχµ+τ-

Marco Zaro, 12/14-05-2015 63

 Time for

Marco Zaro, 12/14-05-2015 64

 NLO exercise

t t ̄ production at NLO
Part 1

• Learn the syntax:

•> tutorial aMCatNLO!
• Generate the code for t t ̄ production at NLO

• Compute the LO and NLO cross-section (run at fixed order)

• Select the analysis analysis_HwU_pp_ttx.o in the
FO_analyse_card to generate histograms 

(need GnuPlot installed)

• In the NLO histograms, which of these variables are described at
the NLO? pT(t), pT(t t),̄ y(t) M(t t),̄ Δɸ(t t)
̄
• What are the histograms with muR=… muF=… for?

Marco Zaro, 12/14-05-2015 65

 NLO exercise

Solution
Part 1

• Learn the syntax:

•> tutorial NLO!
• Generate the code for t t ̄ production at NLO

•> generate p p > t t~ [QCD]!
The current model sm does not allow to generate loop corrections of type QCD.
MG5_aMC now loads 'loop_sm'.
import model loop_sm
. . .
INFO: Generating FKS-subtracted matrix elements for born process: g g > t t~ [ QCD ] (1 / 9)
. . .

•> output my_ttbar_nlo!

• Compute the LO and NLO cross-section

•> launch

Marco Zaro, 12/14-05-2015 66

 NLO exercise

Solution
INFO: ************************************************************
* *
Part 1
* W E L C O M E to M A D G R A P H 5 *

• Learn the syntax:

* a M C @ N L O *
* *
* * * *

•> tutorial NLO!

*
*
* * *
* * * * 5 * * * *
* *
*

• Generate the code for t t ̄ production at NLO

* * * * * *
* * * *
* *

•> *
generate p p > t t~ [QCD]!
*
* allow
The current model sm does not
VERSION 2.2.1

ThetoMadGraph5_aMC@NLO Development
generate loop corrections
2014-09-25

Team
of type - Find us at
QCD.
*
*
*
MG5_aMC now loads 'loop_sm'.* http://amcatnlo.cern.ch *
import model loop_sm * *
. . . * Type 'help' for in-line help. *
INFO: Generating FKS-subtracted
* matrix elements for born process: g g > t t~ [ QCD ] (1 / 9)
*
. . . ************************************************************

•> output my_ttbar_nlo!launch auto

The following switches determine which operations are executed:

• Compute the LO and NLO cross-section

1 Perturbative order of the calculation:
2 Fixed order (no event generation and no MC@[N]LO matching):
3 Shower the generated events:
order=NLO
fixed_order=OFF
shower=ON
•> launch 4 Decay particles with the MadSpin module:
Either type the switch number (1 to 4) to change its default setting,
madspin=OFF

or set any switch explicitly (e.g. type 'order=LO' at the prompt)

Type '0', 'auto', 'done' or just press enter when you are done.
[0, 1, 2, 3, 4, auto, done, order=LO, order=NLO, ... ][60s to answer]
> fixed_order=ON
> order=LO (for LO run)

Marco Zaro, 12/14-05-2015 66

 NLO exercise

Solution
INFO: ************************************************************
* *
Part 1
* W E L C O M E to M A D G R A P H 5 *

• Learn the syntax:

*
INFO: a M C @ N L O *
* Final results and run summary: *
* * * *
Process p p > t t~ [QCD]
•> tutorial NLO!
*
* Run at p-p
*
collider (6500
* *
+ 6500
* * * * 5 * * * *
*
GeV)
*
*

• Generate the code for t t ̄ production at NLO

* Total cross-section:
* 6.871e+02
* * * +- 5.9e+00 pb *
* Ren. and fac. scale * uncertainty: +9.7% * -11.7% *
* *

•> * The results

INFO:
generate p p > t t~ [QCD]!
files
* allow
The current model sm does not
VERSION
of 2.2.1
this run and the HwU2014-09-25

ThetoMadGraph5_aMC@NLO Development
generate loop corrections
and GnuPlot
* with the plots have been saved in /Users/marcozaro/ *
Team
of type - Find us at
QCD.
*

*
Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/Events/run_01
MG5_aMC now loads 'loop_sm'.* http://amcatnlo.cern.ch *
import model loop_sm ! * *
. . . INFO:
* Type 'help' for in-line help. *
INFO: Generating FKS-subtracted
* Final
matrix results
elementsand
for run
bornsummary:
process: g g > t t~ [ QCD ] (1 / 9)
*
. . . Process p p > t t~ [QCD]
************************************************************

•> output my_ttbar_nlo! Run at

launch
TheTotal
autop-p collider (6500 + 6500 GeV)
following switches determine
cross-section: which operations
4.622e+02 +- 2.2e+00 arepb
executed:

• Compute the LO and NLO cross-section

1 Ren.
Perturbative
and fac.order
scaleofuncertainty:
the calculation:

3 Shower the generated events:

+29.8% -22.3%
2 Fixed order (no event generation and no MC@[N]LO matching):
order=NLO
fixed_order=OFF
shower=ON
•> launch INFO: The results of this run and the HwU and GnuPlot
4 Decay particles with the MadSpin module:
files Either
with the plots have been saved in /Users/marcozaro/
type the switch number (1 to 4) to change its default setting,
Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/Events/run_02_LO
madspin=OFF

or set any switch explicitly (e.g. type 'order=LO' at the prompt)

Type '0', 'auto', 'done' or just press enter when you are done.
[0, 1, 2, 3, 4, auto, done, order=LO, order=NLO, ... ][60s to answer]
> fixed_order=ON
> order=LO (for LO run)

Marco Zaro, 12/14-05-2015 66

 NLO exercise

Solution
Part 1

• Select the analysis analysis_HwU_pp_ttx in the
FO_analyse_card to generate histograms

•> launch my_ttbar_nlo!
The following switches determine which operations are executed:
1 Perturbative order of the calculation: order=NLO
2 Fixed order (no event generation and no MC@[N]LO matching): fixed_order=ON
3 Shower the generated events: shower=OFF
4 Decay particles with the MadSpin module: madspin=OFF
Either type the switch number (1 to 4) to change its default setting,
or set any switch explicitly (e.g. type 'order=LO' at the prompt)
Type '0', 'auto', 'done' or just press enter when you are done.
[0, 1, 2, 3, 4, auto, done, order=LO, order=NLO, ... ][60s to answer]
>
INFO: will run in mode: NLO
Do you want to edit a card (press enter to bypass editing)?
1 / param : param_card.dat
2 / run : run_card.dat
3 / FO_analyse : FO_analyse_card.dat
you can also
- enter the path to a valid card or banner.
- use the 'set' command to modify a parameter directly.
The set option works only for param_card and run_card.
Type 'help set' for more information on this command.
- call an external program (ASperGE/MadWidth/...).
Type 'help' for the list of available command
[0, done, 1, param, 2, run, 3, FO_analyse, enter path][60s to answer]

Marco Zaro, 12/14-05-2015 67

 NLO exercise

#######################################################################
#
# This file contains the settings for analyses to be linked to aMC@NLO
# fixed order runs. Analyse files are meant to be put (or linked)

Solution
# inside <PROCDIR>/FixedOrderAnalysis/ (<PROCDIR> is the name of the
# exported process directory). See the
# <PROCDIR>/FixedOrderAnalysis/analysis_template.f file for details on
# how to write your own analysis.
#
#######################################################################
#
Part 1
# Analysis format. Can either be 'topdrawer', 'root', 'HwU' or 'none'.

• Select the analysis analysis_HwU_pp_ttx in the

# When choosing HwU, it comes with a GnuPlot wrapper. When choosing
# topdrawer, the histogramming package 'dbook.f' is included in the
# code, while when choosing root the 'rbook_fe8.f' and 'rbook_be8.cc'
# are included. If 'none' is chosen, all the other entries below have
FO_analyse_card to generate histograms

# to be set empty.
FO_ANALYSIS_FORMAT = HwU

•> #
launch my_ttbar_nlo!
# Needed extra-libraries (FastJet is already linked):
FO_EXTRALIBS
The following switches =
determine which operations are executed:
#
1 Perturbative order of the calculation: order=NLO
# (Absolute) path to the extra libraries. Directory names should be
2 Fixed order (no event generation and no MC@[N]LO matching): fixed_order=ON
# separated by white spaces.
3 Shower the generated events: shower=OFF
FO_EXTRAPATHS =
4 Decay particles with the MadSpin module: madspin=OFF
#
Either type the switch number (1 to 4) to change its default setting,
# (Absolute) path(e.g.
to the dirs containing
at header files needed by the
or set any switch explicitly type 'order=LO' the prompt)
# libraries
'done' (e.g. C++press
header files):
Type '0', 'auto', or just enter when you are done.
FO_INCLUDEPATHS =
[0, 1, 2, 3, 4, auto, done, order=LO, order=NLO, ... ][60s to answer]
#
>
INFO: will run #inUser's
mode: analysis (to be put in the <PROCDIR>/FixedOrderAnalysis/
NLO
Do you want to #edit
directory). Pleaseenter
use .o
to as extension and white spaces to separate
a card (press bypass editing)?
# files.
1 / param : param_card.dat
FO_ANALYSE = analysis_td_template.o
2 / run : run_card.dat
# : FO_analyse_card.dat
3 / FO_analyse
you can also #
- enter the ## When
to linking
a valid with
card root, the following settings are a working
path or banner.
## example on lxplus (CERN). When using this, comment out the lines
- use the 'set' command to modify a parameter directly.
## above andonly
replace <PATH_TO_ROOT> with the physical path to root,
The set option works for param_card and run_card.
Type 'help##set'
e.g.for
/afs/cern.ch/sw/lcg/app/releases/ROOT/5.34.11/x86_64-slc6-gcc46-dbg/root/
more information on this command.
#FO_ANALYSIS_FORMAT = root
- call an external program (ASperGE/MadWidth/...).
#FO_EXTRALIBS = Core Cint Hist Matrix MathCore RIO dl Thread
Type 'help' for the list of available command
#FO_EXTRAPATHS = <PATH_TO_ROOT>/lib
[0, done, 1, param, 2, run, 3, FO_analyse, enter path][60s to answer]
#FO_INCLUDEPATHS = <PATH_TO_ROOT>/include
#FO_ANALYSE = analysis_root_template.o
Marco Zaro, 12/14-05-2015 67
 NLO exercise

#######################################################################
#
# This file contains the settings for analyses to be linked to aMC@NLO
# fixed order runs. Analyse files are meant to be put (or linked)

Solution
# inside <PROCDIR>/FixedOrderAnalysis/ (<PROCDIR> is the name of the
# exported process directory). See the
# <PROCDIR>/FixedOrderAnalysis/analysis_template.f file for details on
# how to write your own analysis.
#
#######################################################################
#
Part 1
# Analysis format. Can either be 'topdrawer', 'root', 'HwU' or 'none'.

• Select the analysis analysis_HwU_pp_ttx in the

# When choosing HwU, it comes with a GnuPlot wrapper. When choosing
# topdrawer, the histogramming package 'dbook.f' is included in the
# code, while when choosing root the 'rbook_fe8.f' and 'rbook_be8.cc'
# are included. If 'none' is chosen, all the other entries below have
FO_analyse_card to generate histograms

# to be set empty.
FO_ANALYSIS_FORMAT = HwU

•> #
launch my_ttbar_nlo!
# Needed extra-libraries (FastJet is already linked):
FO_EXTRALIBS
The following switches =
determine which operations are executed:
#
1 Perturbative order of the calculation: order=NLO
# (Absolute) path to the extra libraries. Directory names should be
2 Fixed order (no event generation and no MC@[N]LO matching): fixed_order=ON
# separated by white spaces.
3 Shower the generated events: shower=OFF
FO_EXTRAPATHS =
4 Decay particles with the MadSpin module: madspin=OFF
#
Either type the switch number (1 to 4) to change its default setting,
# (Absolute) path(e.g.
to the dirs containing
at header files needed by the
or set any switch explicitly type 'order=LO' the prompt)
# libraries
'done' (e.g. C++press
header files):
Type '0', 'auto', or just enter when you are done.
FO_INCLUDEPATHS =
[0, 1, 2, 3, 4, auto, done, order=LO, order=NLO, ... ][60s to answer]
#
>
INFO: will run #inUser's
mode: analysis (to be put in the <PROCDIR>/FixedOrderAnalysis/
NLO
Do you want to #edit
directory). Pleaseenter
use .o
to as extension and white spaces to separate
a card (press bypass editing)?
# files.
1 / param : param_card.dat
2 / run FO_ANALYSE analysis_HwU_pp_ttx.o
= analysis_td_template.o
: run_card.dat
# : FO_analyse_card.dat
3 / FO_analyse
you can also #
- enter the ## When
to linking
a valid with
card root, the following settings are a working
path or banner.
## example on lxplus (CERN). When using this, comment out the lines
- use the 'set' command to modify a parameter directly.
## above andonly
replace <PATH_TO_ROOT> with the physical path to root,
The set option works for param_card and run_card.
Type 'help##set'
e.g.for
/afs/cern.ch/sw/lcg/app/releases/ROOT/5.34.11/x86_64-slc6-gcc46-dbg/root/
more information on this command.
#FO_ANALYSIS_FORMAT = root
- call an external program (ASperGE/MadWidth/...).
#FO_EXTRALIBS = Core Cint Hist Matrix MathCore RIO dl Thread
Type 'help' for the list of available command
#FO_EXTRAPATHS = <PATH_TO_ROOT>/lib
[0, done, 1, param, 2, run, 3, FO_analyse, enter path][60s to answer]
#FO_INCLUDEPATHS = <PATH_TO_ROOT>/include
#FO_ANALYSE = analysis_root_template.o
Marco Zaro, 12/14-05-2015 67
 NLO exercise:

solution
• The HwU (Histogram with Uncertainties) format

##& xmin & xmax & central value & dy & delta_mu_min @aux & delta_mu_max @aux & muR=1.00 muF=1.00 & muR=1.00 muF=2.00 &

muR=0.50 muF=0.50
!


muR=1.00 muF=0.50 & muR=2.00 muF=1.00 & muR=2.00 muF=2.00 & muR=2.00 muF=0.50 & muR=0.50 muF=1.00 & muR=0.50 muF=2.00 &

<histogram> 50 "tt pt |X_AXIS@LIN |Y_AXIS@LOG"

+0.0000000e+00 +2.0000000e+00 -1.0242367e+03 +2.5047252e+01 -1.7206530e+03 -6.0160203e+02 -1.0242367e+03
-9.0715087e+02 -1.1432407e+03 -6.8421704e+02 -6.0160203e+02 -7.6882229e+02 -1.5496422e+03 -1.3802509e+03
-1.7206530e+03
+2.0000000e+00 +4.0000000e+00 +4.9088904e+02 +2.0297264e+01 +3.4493531e+02 +7.1188196e+02 +4.9088904e+02
+4.5019210e+02 +5.3086979e+02 +3.7613186e+02 +3.4493531e+02 +4.0679297e+02 +6.5832080e+02 +6.0377117e+02
+7.1188196e+02
+4.0000000e+00 +6.0000000e+00 +2.2787754e+02 +2.3122314e+01 +1.5999659e+02 +3.3086836e+02 +2.2787754e+02
+2.0857157e+02 +2.4714205e+02 +1.7482611e+02 +1.5999659e+02 +1.8963760e+02 +3.0513912e+02 +2.7932554e+02
+3.3086836e+02
+6.0000000e+00 +8.0000000e+00 +1.7671803e+02 +9.5392210e+00 +1.2453269e+02 +2.5575724e+02 +1.7671803e+02
+1.6227348e+02 +1.9111959e+02 +1.3562893e+02 +1.2453269e+02 +1.4669918e+02 +2.3651862e+02 +2.1720764e+02
+2.5575724e+02
+8.0000000e+00 +1.0000000e+01 +1.2311654e+02 +7.1903869e+00 +8.6399100e+01 +1.7898773e+02 +1.2311654e+02
+1.1261446e+02 +1.3369767e+02 +9.4461506e+01 +8.6399100e+01 +1.0258866e+02 +1.6483914e+02 +1.5078780e+02
+1.7898773e+02
+1.0000000e+01 +1.2000000e+01 +7.8022445e+01 +1.0748137e+01 +5.4873577e+01 +1.1315020e+02 +7.8022445e+01
+7.1570742e+01 +8.4452355e+01 +5.9823787e+01 +5.4873577e+01 +6.4760050e+01 +1.0454718e+02 +9.5909144e+01
+1.1315020e+02
+1.2000000e+01 +1.4000000e+01 +6.1770611e+01 +3.2903213e+00 +4.3437593e+01 +8.9537046e+01 +6.1770611e+01

Marco Zaro, 12/14-05-2015 68

 NLO exercise

Solution
tt pt
tt pt,
3 1 scale variation
10

102
σ per bin [pb]

MadGraph5_aMC@NLO
101

0.8
0.6
0.4
#1 rel.unc.

0.2
0
-0.2
-0.4
Relative uncertainties
-0.6
0 20 40 60 80 100

Marco Zaro, 12/14-05-2015 69

 NLO exercise

Solution
Part 1

• In the NLO histograms, which of these variables are described at
the NLO? pT(t), pT(t t),̄ y(t) M(t t),̄ Δɸ(t t)
̄
• Some of these variables are trivial at LO, because of 2→2
kinematics

• t and t ̄ are always back to back:

d! /d (tt̄) = ( ⇡)
d! /dpT (tt̄) = (pT 0)
• pT(t t)̄ and Δɸ(t t)̄ are non-trivial if the cross-section is at least at
NLO: they are effectively described with LO accuracy

• The other variables are described at NLO

Marco Zaro, 12/14-05-2015 70

 NLO exercise

Solution
Part 1

• What are the histograms with muR=… muF=… for?

• QCD master formula

XZ

 (pp ! tt̄) = dx1 dx2 fa (x1 , µF )fb (x2 , µF ) ⇥ ˆ (ab ! tt̄)
or better
ab
XZ
! (pp ! tt̄) = dx1 dx2 fa (x1 , µF )fb (x2 , µF ) ⇥ ˆ (ab ! tt̄; µF , µR , ↵S (µR ))
• What are μF/R?
ab

• They are arbitrary scales needed to renormalise the strong coupling and
to reabsorb initial state IR-divergences in PDFs, chosen to be of the order
of the hard scattering scales (sum of masses, pT, …)

• The all-order cross-section is independent of the choice of μF/R

• At NkLO, the dependence is of Nk+1LO

• Computing the cross-section with different scales can be a way to
estimate uncertainties due to missing higher orders

• How much scales are varied is arbitrary, usually in the range [0.5, 2]
Marco Zaro, 12/14-05-2015 71
 Scale uncertainties

Marco Zaro, 12/14-05-2015 72

 Scale uncertainties
• Look at the LO and NLO cross-section we have just computed

• Values with different scales are computed on the fly and the
envelope is taken INFO:
Final results and run summary:
Process p p > t t~ [QCD]
Run at p-p collider (6500 + 6500 GeV)
Total cross-section: 6.871e+02 +- 5.9e+00 pb
Ren. and fac. scale uncertainty: +9.7% -11.7%

INFO: The results of this run and the TopDrawer file with
the plots have been saved in /Users/marcozaro/Physics/
MadGraph/2.2.3new/my_tt_nlo_qcd/Events/run_01
!
INFO:
Final results and run summary:
Process p p > t t~ [QCD]
Run at p-p collider (6500 + 6500 GeV)
Total cross-section: 4.622e+02 +- 2.2e+00 pb
Ren. and fac. scale uncertainty: +29.8% -22.3%

INFO: The results of this run and the TopDrawer file with
the plots have been saved in /Users/marcozaro/Physics/
MadGraph/2.2.3new/my_tt_nlo_qcd/Events/run_02_LO

Marco Zaro, 12/14-05-2015 72

 Scale uncertainties
• Look at the LO and NLO cross-section we have just computed

• Values with different scales are computed on the fly and the
envelope is taken
• Typically LO has larger scale uncertainties

Marco Zaro, 12/14-05-2015 72

 Scale uncertainties
• Look at the LO and NLO cross-section we have just computed

• Values with different scales are computed on the fly and the
envelope is taken
• Typically LO has larger scale uncertainties
• To have scale uncertainties for distributions, one must fill one
histogram per scale choice, and then take the envelope

Marco Zaro, 12/14-05-2015 72

 Scale uncertainties
• Look at the LO and NLO cross-section we have just computed

• Values with different scales are computed on the fly and the
envelope is taken
• Typically LO has larger scale uncertainties
• To have scale uncertainties for distributions, one must fill one
histogram per scale choice, and then take the envelope
• The same is possible for PDF uncertainties

Marco Zaro, 12/14-05-2015 72

 Scale uncertainties
tt inv m tt inv m
tt inv m, tt inv m,
1 scale variation 1 scale variation

102
102

NLO
 LO

σ per bin [pb]

σ per bin [pb]

~10% unc. ~30% unc.
101

101

MadGraph5_aMC@NLO

MadGraph5_aMC@NLO
100

100

0.5
0.2 0.4
0.3
0.1 0.2
#1 rel.unc.

#1 rel.unc.
0.1
0 0
-0.1
-0.1
-0.2
-0.2 Relative uncertainties -0.3 Relative uncertainties
-0.4
400 500 600 700 800 900 1000 400 500 600 700 800 900 1000

Marco Zaro, 12/14-05-2015 73

 Scale uncertainties

pT(t t)̄ histogram from NLO run tt pt

tt pt,
103 1 scale variation

effectively LO

102 ~30-40% unc.
σ per bin [pb]

MadGraph5_aMC@NLO
101

0.8
0.6
0.4
#1 rel.unc.

0.2
0
-0.2
-0.4
Relative uncertainties
-0.6
0 20 40 60 80 100

Marco Zaro, 12/14-05-2015 74

 NLO exercise

t t ̄ production at NLO
Part 2

• Generate a NLO event sample to be showered by Pythia6Q

• Shower and analyse it with the py6an_HwU_pp_ttx.o analysis (to be specified
in the shower_card)

• The histogramming routine (HwU.o) must also be added to the analysis files in the
shower_card (Hint: you can shower an existing run with ./bin/shower run_xx)

• Use MadSpin to generate a di-leptonic (into muons) decayed sample

• Re-analyse the decayed and undecided sample with the
py6an_HwU_pp_lplm.o analysis and check the lepton pair pT

• The analysis (in MCatNLO/PYAnalyzer/py6an_HwU_pp_lplm.f)has to be slightly
modified:

• IORI.LE.10 → IORI.LE.20 at lines 186, 190

• To tell Pythia to perform di-leptonic decays, add these lines in the shower_card
(‘Decay channels’ block; antiparticles are decayed as particles)

• DM_1 = 6 > 24 5 @1d0 @100

DM_2 = 24 > 14 -13 @1d0 @100

Marco Zaro, 12/14-05-2015 75

 NLO exercise

Solution
Part 2

• Generate a NLO event sample to be showered by Pythia6Q

• Shower it with the mcatnlo_pyan_pp_ttx analysis (to be specified in the
shower_card)

• cd my_ttbar_nlo!
• ./bin/aMCatNLO!
• > launch!
• > fixed_order=OFF!
• > shower=ON!
• Edit run_card

• Edit shower_card

Marco Zaro, 12/14-05-2015 76

 NLO exercise

Solution
Part 2
# PDF choice: this automatically fixes also alpha_s(MZ) and its evol.

•
*

Generate a NLO event sample to be showered by Pythia6Q

#**********************************************************************
*

• Shower it with the mcatnlo_pyan_pp_ttx analysis (to be specified in the

nn23nlo = pdlabel ! PDF set
244600 = lhaid ! if pdlabel=lhapdf, this is the lhapdf number
#**********************************************************************

shower_card)

*
# Include the NLO Monte Carlo subtr. terms for the following parton
*

• cd my_ttbar_nlo! # shower (HERWIG6 | HERWIGPP | PYTHIA6Q | PYTHIA6PT | PYTHIA8)

*

• ./bin/aMCatNLO!
# WARNING: PYTHIA6PT works only for processes without FSR!!!!
*
#**********************************************************************

• > launch! *
PYTHIA6Q = parton_shower
#**********************************************************************

• > fixed_order=OFF! *
# Renormalization and factorization scales

• > shower=ON! *
# (Default functional form for the non-fixed scales is the sum of

•
*

Edit run_card
# the transverse masses of all final state particles and partons. This
*

•
# can be changed in SubProcesses/set_scales.f)

Edit shower_card *
#**********************************************************************
*
F = fixed_ren_scale ! if .true. use fixed ren scale
F = fixed_fac_scale ! if .true. use fixed fac scale
91.188 = muR_ref_fixed ! fixed ren reference scale
91.188 = muF1_ref_fixed ! fixed fact reference scale for pdf1
91.188 = muF2_ref_fixed ! fixed fact reference scale for pdf2
#**********************************************************************
*
# Renormalization and factorization scales (advanced and NLO options)
*
#**********************************************************************
*

Marco Zaro, 12/14-05-2015 76

 NLO exercise

Solution
Part 2
# PDF choice: this automatically fixes also alpha_s(MZ) and its evol.

•
*

Generate a NLO event sample to be showered by Pythia6Q

#**********************************************************************
*

• Shower it with the mcatnlo_pyan_pp_ttx analysis (to be specified in the

nn23nlo = pdlabel ! PDF set
244600 = lhaid ! if pdlabel=lhapdf, this is the lhapdf number
#**********************************************************************

shower_card)

*
# Include the NLO Monte Carlo subtr. terms for the following parton
*

• cd my_ttbar_nlo! # shower (HERWIG6 | HERWIGPP | PYTHIA6Q | PYTHIA6PT | PYTHIA8)

*

• ./bin/aMCatNLO!
# WARNING: PYTHIA6PT works only for processes without FSR!!!!
*
#**********************************************************************

• > launch! *
PYTHIA6Q = parton_shower
#**********************************************************************

• > fixed_order=OFF! *
# Renormalization and factorization scales

• > shower=ON! *
#***********************************************************************
# (Default functional form for the non-fixed scales is the sum of
# Extra Libraries/analyses *

•
*

Edit run_card

# The following lines need to be changed if the user does
# the transverse masses of all final state particles and partons. This not want to *
* # create a StdHEP/HepMC file, but to directly run an own analysis (to *

•
# can# be
be changed
placed in
in HWAnalyzer or analogous MCatNLO subfolders).
SubProcesses/set_scales.f) *

Edit shower_card * # Please use files in those folders as examples.

#***********************************************************************
#**********************************************************************
EXTRALIBS = # Extra-libraries (not LHAPDF)
*

*
F = fixed_ren_scale ! if .true.# Default:
use fixed "stdhep Fmcfio"
ren scale
F # PYTHIA > 8.200
= fixed_fac_scale ! if .true. use fixed fac scalemay require library dl
EXTRAPATHS
91.188 = ../lib
= muR_ref_fixed # Path to the extra-libraries
! fixed ren reference scale
91.188 = muF1_ref_fixed # Default:
! fixed fact reference"../lib"
scale for pdf1
INCLUDEPATHS
91.188 =
= muF2_ref_fixed # Path to header
! fixed fact reference scale filesforneeded
pdf2 by c++
# Dir names separated by
#********************************************************************** white spaces
* ANALYSE = mcatnlo_pyan_pp_ttx.o mcatnlo_hbook_gfortran8.o
# Renormalization and factorization scales# routines (please
(advanced and use
NLO .o as extension
options)
* # and use spaces to separate files)
#**********************************************************************
*

Marco Zaro, 12/14-05-2015 76

 NLO exercise

Solution
Part 2
# PDF choice: this automatically fixes also alpha_s(MZ) and its evol.

•
*

Generate a NLO event sample to be showered by Pythia6Q

#**********************************************************************
*
Summary:
Process p p > t t~ [QCD]

• Shower it with the mcatnlo_pyan_pp_ttx analysis (to be specified in the

nn23nlo = pdlabel ! PDF set
Run
244600
at p-p
= lhaid
collider (6500 + 6500 GeV)
! if pdlabel=lhapdf, this is the lhapdf number
Total cross-section: 6.772e+02 +- 2.1e+00 pb
#**********************************************************************
Ren. and fac. scale uncertainty: +11.5% -13.0%
shower_card)

*
Number
# Include the ofNLOevents generated:
Monte Carlo 100000
subtr. terms for the following parton
* Parton shower to be used: PYTHIA6Q
• cd my_ttbar_nlo! # shower
*
(HERWIG6
Fraction of |negative
HERWIGPP |weights:
Total running time : 6m 58s
PYTHIA6Q |0.20
PYTHIA6PT | PYTHIA8)

• ./bin/aMCatNLO!
# WARNING: PYTHIA6PT works only for processes without FSR!!!!
*
INFO: The /Users/marcozaro/Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/Events/
#**********************************************************************

• > launch! run_12/events.lhe.gz

*
. . . PYTHIA6Q =
file has been generated.
parton_shower
#**********************************************************************
INFO: Preparing MCatNLO run
• > fixed_order=OFF! *
INFO: Compiling MCatNLO for PYTHIA6Q...
# Renormalization and factorization scales
INFO: ... done

• > shower=ON! *
INFO:
INFO: #
Showering
# (Default
(Running
Extra in
events...
#***********************************************************************
functional form for the non-fixed scales is the sum of
/Users/marcozaro/Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/
Libraries/analyses *

•
*

Edit run_card

# The following
MCatNLO/RUN_PYTHIA6Q_3) lines need to be changed if the user
# the transverse masses of all final state particles and partons. This does not want to *
INFO:
* # Idle:
create 0,a StdHEP/HepMC
Running: file, but to directly
1, Completed: 0 [ run an own time:
current analysis (to ]*
12h32

•
#
# can bebe placed
changed in
in HWAnalyzer or analogous
SubProcesses/set_scales.f) MCatNLO subfolders). *
INFO: Idle: 0, Running: 0, folders
Completed: 1 [ 2m 35s ]
Edit shower_card *
INFO:
INFO:
# Please use files in those as examples.
Idle: 0, Running: 0, Completed: 0 [ current time: 12h34 ]
#***********************************************************************
#**********************************************************************
The file /Users/marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01/
EXTRALIBS = # Extra-libraries (not LHAPDF)
*

*
plot_PYTHIA6Q_1_0.HwU
F = fixed_ren_scale been
has generated,
# Default:
! if .true. with
use fixed renhistograms
"stdhep Fmcfio"
scale in the HwU and
GnuPlot
F formats, obtained by
= fixed_fac_scale ! if # PYTHIA
showering
.true. usethe >
fixed8.200 may
parton-level
fac scalerequire library
file /Users/dl
EXTRAPATHS
91.188 = ../lib
= muR_ref_fixed # Path to the
! fixed ren reference scaleextra-libraries
marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01/events.lhe.gz with
91.188 = muF1_ref_fixed ! fixed # Default:
fact "../lib"
reference scale for pdf1
PYTHIA6Q.
INCLUDEPATHS
91.188 =
= muF2_ref_fixed # Path
! fixed fact to header
reference files
scale forneeded
pdf2 by c++
INFO: Run complete # Dir names separated by white spaces
#**********************************************************************
* ANALYSE = mcatnlo_pyan_pp_ttx.o mcatnlo_hbook_gfortran8.o
# Renormalization and factorization scales # routines (please
(advanced and use
NLO .o as extension
options)
* # and use spaces to separate files)
#**********************************************************************
*

Marco Zaro, 12/14-05-2015 76

 NLO exercise

tt pt
Solution
Part 2
# PDF choice: this automatically fixes also alpha_s(MZ) tt pt, NOCUT and its evol.

•
NOCUT scale variation
* CUT

Generate a NLO event sample to be showered by Pythia6Q

#**********************************************************************
*
Summary:
Process p p > t t~ [QCD]
CUT scale variation

• Shower it with the mcatnlo_pyan_pp_ttx analysis (to be specified in the

nn23nlo = pdlabel ! PDF set
102
244600
Run at p-p
= lhaid
collider (6500 + 6500 GeV)
! if pdlabel=lhapdf, this is the lhapdf number
Total cross-section: 6.772e+02 +- 2.1e+00 pb
#**********************************************************************
Ren. and fac. scale uncertainty: +11.5% -13.0%
shower_card)

*
# IncludeNumber the ofNLOevents generated:
Monte Carlo 100000
subtr. terms for the following parton
σ per bin [pb]

* Parton shower to be used: PYTHIA6Q

• cd my_ttbar_nlo! # shower
*
(HERWIG6
Fraction of |negative
HERWIGPP |weights:
Total running time : 6m 58s
PYTHIA6Q |0.20PYTHIA6PT | PYTHIA8)

• ./bin/aMCatNLO!
# WARNING: PYTHIA6PT works only for processes without FSR!!!!

MadGraph5_aMC@NLO
*
101 INFO: The /Users/marcozaro/Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/Events/
#**********************************************************************

• > launch! run_12/events.lhe.gz

*
. . . PYTHIA6Q =
file has been generated.
parton_shower
#**********************************************************************
INFO: Preparing MCatNLO run
• > fixed_order=OFF! *
INFO: Compiling MCatNLO for PYTHIA6Q...
# Renormalization and factorization scales
INFO: ... done

• > shower=ON! *
INFO:
INFO: #
Showering
# (Default
(Running
Extra in
events...
#***********************************************************************
functional form for the non-fixed scales is the sum of
/Users/marcozaro/Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/
Libraries/analyses *

•
0.3 *

Edit run_card

# The following
MCatNLO/RUN_PYTHIA6Q_3) lines need to be changed if the
# the transverse masses of all final state particles and partons. This user does not want to *
NOCUT rel.unc.

0.2
INFO:
* # Idle:
create 0,a StdHEP/HepMC
Running: file, but to directly
1, Completed: 0 [ run an own time:
current analysis (to ]*
12h32

•
0.1
# can be# be placed
changed in
in HWAnalyzer or analogous
SubProcesses/set_scales.f) MCatNLO subfolders). *
INFO: Idle: 0, Running: 0, folders
Completed: 1 [ 2m 35s ]
Edit shower_card # Please use files in those as examples. *
0
*
-0.1 INFO: Idle: 0, Running: 0, Completed: 0 [ current time: 12h34 ]
#***********************************************************************
#**********************************************************************
-0.2
INFO:
* The file /Users/marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01/
EXTRALIBS = # Extra-libraries (not LHAPDF)
Relative uncertainties
-0.3
plot_PYTHIA6Q_1_0.HwU
F = fixed_ren_scale been
has generated,
! if .true.# Default:
use fixed with renhistograms
"stdhep Fmcfio"
scale in the HwU and
GnuPlot
F formats, obtained by
= fixed_fac_scale ! if # PYTHIA
showering
.true. usethe >
fixed 8.200 may
parton-level
fac scale require library
file /Users/dl
0.95 EXTRAPATHS = ../lib # Path to
marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01/events.lhe.gz
91.188 = muR_ref_fixed the
! fixed ren reference scale extra-libraries with
91.188 = muF1_ref_fixed ! fixed # Default:
fact reference "../lib"
scale for pdf1
0.9
PYTHIA6Q.
(2)/NOCUT

91.188INCLUDEPATHS =
= muF2_ref_fixed # Path
! fixed fact to header
reference files
scale forneeded
pdf2 by c++
0.85 INFO: Run complete # Dir names separated by white spaces
#**********************************************************************
0.8 ANALYSE = mcatnlo_pyan_pp_ttx.o mcatnlo_hbook_gfortran8.o
*
0.75 # Renormalization and factorization scales # routines
(advanced (pleaseand use
NLO .o as extension
options)
0.7
(2)/NOCUT
* # and use spaces to separate files)
0
#**********************************************************************
20 40 60 80 100
*

Marco Zaro, 12/14-05-2015 76

 The events
<initrwgt>
<weightgroup type='scale_variation' combine='envelope'>
<weight id='1001'> muR=0.10000E+01 muF=0.10000E+01 </weight>
<weight id='1002'> muR=0.10000E+01 muF=0.20000E+01 </weight>
<weight id='1003'> muR=0.10000E+01 muF=0.50000E+00 </weight>
<weight id='1004'> muR=0.20000E+01 muF=0.10000E+01 </weight>
• Each event keeps information about
<weight id='1005'> muR=0.20000E+01 muF=0.20000E+01 </weight>
<weight id='1006'> muR=0.20000E+01 muF=0.50000E+00 </weight>
scale variations

<weight id='1007'> muR=0.50000E+00 muF=0.10000E+01 </weight>
<weight id='1008'> muR=0.50000E+00 muF=0.20000E+01 </weight>
• To obtain scale uncertainties use the
<weight id='1009'> muR=0.50000E+00 muF=0.50000E+00 </weight>
</weightgroup>
extra weights to fill histograms and
</initrwgt>
</header> take the envelope
<init>
2212 2212 0.65000000E+04 0.65000000E+04 -1 -1 244600 244600 -4 1
0.68147533E+03 0.22760274E+01 0.11811897E+04 0
</init>
<event>
4 0 -.11811897E+04 0.68991465E+03 0.75467716E-02 0.11800000E+00
21 -1 0 0 501 502 0.00000000E+00 0.00000000E+00 0.16695776E+03 0.16695776E+03 0.00000000E+00 0.0000E+00 0.9000E+01
21 -1 0 0 502 503 -.00000000E+00 -.00000000E+00 -.83539498E+03 0.83539498E+03 0.00000000E+00 0.0000E+00 0.9000E+01
6 1 1 2 501 0 -.87405313E+02 -.30435858E+03 -.46344397E+03 0.58735266E+03 0.17300000E+03 0.0000E+00 0.9000E+01
-6 1 1 2 0 503 0.87405313E+02 0.30435858E+03 -.20499324E+03 0.41500008E+03 0.17300000E+03 0.0000E+00 0.9000E+01
#aMCatNLO 1 5 3 3 2 0.21343976E+03 0.35860250E+02 9 0 0 0.10000001E+01 0.15353083E+01 0.66887201E+00 0.00E+00 0.0E+00
<rwgt>
<wgt id='1001'> -.11812E+04 </wgt>
<wgt id='1002'> -.10571E+04 </wgt>
<wgt id='1003'> -.13263E+04 </wgt>
<wgt id='1004'> -.88285E+03 </wgt>
<wgt id='1005'> -.79006E+03 </wgt>
<wgt id='1006'> -.99128E+03 </wgt>
<wgt id='1007'> -.16151E+04 </wgt>
<wgt id='1008'> -.14453E+04 </wgt>
<wgt id='1009'> -.18135E+04 </wgt>
</rwgt>
</event>

Marco Zaro, 12/14-05-2015 77

 NLO exercise

Solution
Part 2

• Use MadSpin to generate a di-leptonic (into muons) decayed sample

• ./bin/aMCatNLO!
• > decay_events run_xx!
• edit the madspin_card

Marco Zaro, 12/14-05-2015 78

 NLO exercise

Solution
Part 2

• Use MadSpin to generate a di-leptonic (into muons) decayed sample

• ./bin/aMCatNLO!
• > decay_events run_xx!
• edit the madspin_card
#************************************************************
#* MadSpin *
#* *
#* P. Artoisenet, R. Frederix, R. Rietkerk, O. Mattelaer *
#* *
#* Part of the MadGraph5_aMC@NLO Framework: *
#* The MadGraph5_aMC@NLO Development Team - Find us at *
#* https://server06.fynu.ucl.ac.be/projects/madgraph *
#* *
#************************************************************
#Some options (uncomment to apply)
#
# set seed 1
# set Nevents_for_max_weigth 75 # number of events for the estimate of the max. weight
# set BW_cut 15 # cut on how far the particle can be off-shell
set max_weight_ps_point 400 # number of PS to estimate the maximum for each event
#
# specify the decay for the final state particles
decay t > w+ b, w+ > mu+ vm
decay t~ > w- b~, w- > mu- vm~
# running the actual code
launch
~

Marco Zaro, 12/14-05-2015 78

 NLO exercise

Solution
Part 2

• Use MadSpin to generate .a. di-leptonic
.
(into muons) decayed sample

• ./bin/aMCatNLO! INFO: MadSpin: Estimate the maximum weight
INFO:

• > decay_events run_xx!

INFO:
INFO:
Estimating the maximum weight
*****************************

•
INFO: Probing the first 139 events
edit the madspin_card
INFO:
INFO:
with 400 phase space points

INFO: Event 1/139 : 0.059s

#*
INFO: Event 6/139 : 0.99s
INFO: Event 11/139 : 1.3s
#************************************************************
MadSpin . . . *
MS
⇣ estimates
⌘
2 *2 2 2
#*
#* |M
P. Artoisenet, R. Frederix, R. Rietkerk,| / |M | > Rand() max |M
INFO: Decaying the events...
P +D P *
INFO: EventO.nbMattelaer
1000 2.6s
P+D! | / |M | P
#* *
#*
#*
#*
Part of the MadGraph5_aMC@NLOINFO:

. . .
Event nb 2000 4.9s
Framework:
INFO: Event nb 3000
The MadGraph5_aMC@NLO Development Team - Find us at
https://server06.fynu.ucl.ac.be/projects/madgraph
*
7s
*
*
with the first events
#* INFO: Decayed events have * been written in /Users/marcozaro/Physics/
#************************************************************
MadGraph/2.2.3new/my_tt_nlo_qcd/Events/run_01/events_decayed.lhe.gz
#Some options (uncomment to apply) INFO: The decayed event file has been moved to the following location:
#
# set seed 1
INFO: /Users/marcozaro/Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/Events/
run_01_decayed_1/events.lhe.gz
# set Nevents_for_max_weigth 75 # number of events for the estimate of the max. weight
# set BW_cut 15 INFO: MadSpin
# cut on how far theDone
particle can be off-shell
set max_weight_ps_point 400 # number of PS to estimate the maximum for each event
#
# specify the decay for the final state particles
decay t > w+ b, w+ > mu+ vm
decay t~ > w- b~, w- > mu- vm~
# running the actual code
launch
~

Marco Zaro, 12/14-05-2015 78

 NLO exercise

Solution
Part 2

• Re-analyse the decayed and undecided sample with the py6an_HwU_pp_lplm
analysis and check the the lepton pair pT

• Re-shower the un-decayed sample

• ./bin/shower run_xx!
• edit the shower_card

• Shower the decayed sample

• ./bin/shower run_xx_decayed_1

Marco Zaro, 12/14-05-2015 79

 NLO exercise

Solution
Part 2

• Re-analyse the decayed and undecided sample with the py6an_HwU_pp_lplm
analysis and check the the lepton pair pT

• Re-shower the un-decayed sample

• ./bin/shower run_xx!
•
# Decay channels *

edit the shower_card

# Write down the decay channels for the resonances, to be performed by *
# the shower. *

• Shower the decayed sample

. . .
DM_1 = 6 > 24 5 @1d0 @100

DM_2 = -6 > -24 -5 @1d0 @100


• ./bin/shower DM_3 = 24 > 14 -13 @1d0 @100


run_xx_decayed_1
DM_4 = —24 > -14 13 @1d0 @100
#***********************************************************************
#***********************************************************************
# Extra Libraries/analyses *
# The following lines need to be changed if the user does not want to *
# create a StdHEP/HepMC file, but to directly run an own analysis (to *
# be placed in HWAnalyzer or analogous MCatNLO subfolders). *
# Please use files in those folders as examples. *
#***********************************************************************
EXTRALIBS = # Extra-libraries (not LHAPDF)
# Default: "stdhep Fmcfio"
# PYTHIA > 8.200 may require library dl
EXTRAPATHS = ../lib # Path to the extra-libraries
# Default: "../lib"
INCLUDEPATHS = # Path to header files needed by c++
# Dir names separated by white spaces
ANALYSE = mcatnlo_pyan_pp_lplm.o mcatnlo_hbook_gfortran8.o
# routines (please use .o as extension
# and use spaces to separate files)

Marco Zaro, 12/14-05-2015 79

 NLO exercise

Solution
Part 2

• Re-analyse the decayed and undecided sample with the py6an_HwU_pp_lplm
analysis and check the the lepton pair pT

• Re-shower the un-decayed sample

• ./bin/shower run_xx!
INFO: Preparing MCatNLO run

•
INFO: Compiling MCatNLO for PYTHIA6Q...
# Decay channels *

edit the shower_card

INFO:
INFO: Showering events...
...# done
Write down the decay channels for the resonances, to be performed by *
# the shower. *

• Shower the decayed sample

MCatNLO/RUN_PYTHIA6Q_3)
INFO: Idle: 0, Running: 1, DM_3
. . .
INFO: (Running in /Users/marcozaro/Physics/MadGraph/2.2.3new/my_tt_nlo_qcd/
DM_1 = 6 > 24 5 @1d0 @100

DM_2 = -6 > -24 -5 @1d0 @100

Completed:
= 24 > 140-13
[ current
@1d0 @100
time: 12h32 ]
• ./bin/shower run_xx_decayed_1
INFO: Idle: 0, Running: 0, DM_4 Completed:
= —24 > -14 13 2m
1 [ @1d035s
@100 ]
INFO: Idle: 0, Running: 0, #***********************************************************************
Completed: 0 [ current time: 12h34 ]
INFO: The file /Users/marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01/
#***********************************************************************
plot_PYTHIA6Q_2_0.HwU has been# generated,
Extra Libraries/analyses
with histograms in the HwU and *
# The following lines need to be changed
GnuPlot formats, obtained by showering the parton-level file /Users/ if the user does not want to *
# create a StdHEP/HepMC file, but to directly run an own analysis (to *
marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01/events.lhe.gz with
# be placed in HWAnalyzer or analogous MCatNLO subfolders). *
PYTHIA6Q. # Please use files in those folders as examples. *
INFO: Run complete #***********************************************************************
. . . EXTRALIBS = # Extra-libraries (not LHAPDF)
INFO: Idle: 0, Running: 1, Completed: 0 [ current time:# Default: 12h32 ] "stdhep Fmcfio"
INFO: Idle: 0, Running: 0, Completed: 1 [ 2m 35s ] # PYTHIA > 8.200 may require library dl
EXTRAPATHS = ../lib # Path to the extra-libraries
INFO: Idle: 0, Running: 0, Completed: 0 [ current time: 12h34 ]
# Default: "../lib"
INFO: INFO: The file /Users/marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/
INCLUDEPATHS = # Path to header files needed by c++
run_01_decayed_1/plot_PYTHIA6Q_1_0.HwU has been generated,# Dir withnames
histograms
separatedin
by white spaces
the HwU and GnuPlot formats, obtained
ANALYSE by =showering the parton-level
mcatnlo_pyan_pp_lplm.o file /
mcatnlo_hbook_gfortran8.o
Users/marcozaro/Physics/MadGraph/2.3.1/ttbar/Events/run_01_decayed_1/
# routines (please use .o as extension
events.lhe.gz with PYTHIA6Q. # and use spaces to separate files)
INFO: Run complete

Marco Zaro, 12/14-05-2015 79

 √
S = 14 TeV) and at the Tevatron run II
the predictions given in this section have b
NLO exercise

default PDF set [9], and by setting mt = 1
Solution
of single-t production, we also reconstruct t
kT -clustering algorithm [10], with dcut = 1
Part 2
procedure all final-state stable hadrons and
• Re-analyse the decayed and undecidedforce
sample
π 0 ’swith
andtheall lowest-lying b-flavoured st
mcatnlo_pyan_pp_lplm analysis are and ordered
check the in the lepton pair
transverse pT
momentum.
+-
pT(l l ) Plot from Frixione, Leanen, Motylinski, Webber, arXiv:hep-ph/0702198
100

10-1
σ per bin [pb]

lp pt, MS
lp pt, PY
MadGraph5_aMC@NLO

-2
10

10-3
0 50 100 150 200

Marco Zaro, 12/14-05-2015 80