MadGraph5_aMC@NLO

higgs anomalous coupling to photon and b quarks

Asked by Rodrigo Capucha

Dear Madgraph experts,

I am trying to estimate the cross section of the process p p > h, h > a b b~ in the SM. For that, I'm using (in Madgraph 2.9.14) the heft model, and I tried the following:

import model heft
generate p p > h, h > a b b~

I've applied some cuts to compare with a value I have found in the literature, such as eta_b/a <= 2.5, deltaR(a,b) > 0.4, pTb > 20 GeV, pTa > 40 GeV, and setting wh = 4.4 MeV and sqrt(s) = 14 TeV, I got a cross section of approximately 1.35 pb, which is nowhere near the cross section value that I found (0.00064pb).

My first question is if I did something obviously wrong. Also, if it's not asking too much, can you reproduce the value that I get for this cross section?

I also got a warning message during the generation of the process, which makes me question even more the number that I got. The message was the following:

"
INFO: fail to reach target 1000
failed to generate enough events. Please follow one of the following suggestions to fix the issue:
  - set in the run_card.dat 'sde_strategy' to 3
  - set in the run_card.dat 'hard_survey' to 1 or 2.
  - reduce the number of requested events (if set too high)
  - check that you do not have -integrable- singularity in your amplitude.
"

I've tried the suggestions proposed (sde_strategy to 1 or 2, 3 doesn't exist, and hard_survey to 1 or 2; the number of events was not too high, I changed it between 500 and 10k events), but I keep getting this message. Is this something I should worry about?

Thank you in advance for your time.

Cheers,
Rodrigo

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

If i do run the following script:

import model heft
generate p p > h, h > a b b~
output
launch
shower=off
set no_parton_cut
set cut_decays T
set etab 2.5
set etaa 2.5
set drab 0.4
set ptb 20
set pta 40
set wh 0.0044

Then the cross-section that I got is:
#***************************************************************************
#
# original cross-section: 0.0007941617752898606
# scale variation: +2.18% -4.68%
# central scheme variation: +9.24e-10% -5.37%
# PDF variation: +3.13% -3.13%
#
# dynamical scheme # 1 : 0.000790363 +2.56% -4.94% # \sum ET
# dynamical scheme # 2 : 0.000790499 +2.55% -4.93% # \sum\sqrt{m^2+pt^2}
# dynamical scheme # 3 : 0.000751537 +5.18% -7.56% # 0.5 \sum\sqrt{m^2+pt^2}
# dynamical scheme # 4 : 0.000794162 +2.18% -4.68% # \sqrt{\hat s}
#***************************************************************************

On the other hand, I do reproduce your number if I do not apply any cuts (so if I keep cut_decays on False)
  === Results Summary for run: run_01 tag: tag_1 ===
     Cross-section : 1.117 +- 0.01239 pb
     Nb of events : 981

Where the lack of events is just related to the soft singularity of the photon which is not regulated.

Cheers,

Olivier

Revision history for this message
Rodrigo Capucha (mrcapsy) said : 		#2

Hi Olivier,

Thank you so much, this solved my problem. The number is not quite what I want, but it's close now, probably there are some additional cuts that I've missed. I've been using Madgraph for a while now, but I never noticed that you had to set an option to true for the cuts to be applied to the decay products as well. Is this also the case when you use MadSpin to decay the particles?

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#3

They are no real way to apply cut with madspin.

If you use spinmode=none (which is valid for the higss case), then you can define a second run_card where you can defined cut that will be applied on the higgs center of mass frame.
Therefore all the rapidity/pt cut will not make sense to use in that card.

And to be complete, if your decay in madspin contains onshell resonance, then in that case, you will also be impacted by the cut_decays flag.

Cheers,

Olivier

Revision history for this message
Rodrigo Capucha (mrcapsy) said : 		#4

Thanks Olivier Mattelaer, that solved my question.