MadGraph5_aMC@NLO

Using model sm versus model sm-no_b_mass

Asked by Gabriel Vian

Dear MadGraph team

I'm working with a process where H --> bb, but in the default card, my team is using, the model "sm-no_b_mass" is being applied to the generation of events.

Researching the differences between the "sm-no_b_mass" model and the "sm" model I found (here https://answers.launchpad.net/mg5amcnlo/+question/453586) that using b mass equals zero is more coherent because the PDFs used by the program assume DGLAP massless equations.

 I would like to know if it is possible to give a non-zero value to the Higgs-Bottom Yukawa coupling while using sm-no_b_mass.

Or if it would not be problematic to use "sm" model, as there might be this above-mentioned incoherence.

Thanks in advance for your assistance.

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

Yes this is doable both at LO and at NLO.
At LO this is simple while this is more complex at NLO (and might depend on the model implementation)

For LO, you just will need to define your own restriction_card (see the associated FAQ)

Cheers,

Olivier

FAQ #2312: “FR Model much slower than build-in MG model. Why and how to fix?”.

Revision history for this message
Gabriel Vian (gabriel-vian-24) said : 		#2

Dear Olivier,

I was not able to understand completely the process of making a restriction card.

1) Would it be possible to do it more easily, like setting a non-zero value for the Yukawa coupling in the param card?

2) Using "sm" model in this case would be problemátic?

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

Well in your case,

take the file restrict-no_b_mass.dat
and change the value of the yukawa to a non zero value and rename the file
to restrict-masslessb_with_yukawa

and then use the model
import model sm-masslessb_with_yukawa

This should be easy enough

Concerning your question 2, I do not have enough information to comment on it.
At LO it is likely a small effect (if you have b in the initial state) this is quite problematic at NLO obviously.

Cheers,

Olivier

Revision history for this message
Gabriel Vian (gabriel-vian-24) said : 		#4

Dear Olivier,

Thank you for your patience and helpful explanation.

In my current work on generating grid packs, we typically use a "run_card," a "process_card," and a "customization_card" for the setup.

Instead of creating a new "restrict-masslessb_with_yukawa" file and importing the model with this restriction, would it be possible to simply set a Yukawa value for the b quark directly in the "customization_card"?

Thank you again for your assistance.

Best regards,
Gabriel

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#5

Hi,

The customaization_card is not something that does exists in MG5aMC. This is something internal to CMS, so I'm not the expert on what you can/cannot do for that file. My understanding is that in that file you can pass "set" command that can edit the param_card/run_card(or any other card) since you pass a run_card.I guess that you are suppose to restrict yourself to the param_card.dat

If that above statement is True, this is too late to put back the yukawa to non zero since at this level the fortran code that evaluate the function does not have any variable associated to such yukawa. (and at that stage, you will not have have those diagram to evaluate).

Conceptually, any deep change to the model should be done within the process_card (if his content corresponds to the file that I guess --this naming is also CMS specific, we do not have/use such filename since 2014 --).
The line that you have to change is really the "import model".
The only possible work-around would be to use afterwards the "customize_model" functionality, but that functionality (to create restriction_file on the flight) is not generic enough to allow it (But this is something that I can consider to update).

Now what you can do is to do your own copy of the SM and treat it as it is handle for BSM model within CMS (in that case please contact the correct group within CMS)

Cheers,

Olivier

Revision history for this message
Gabriel Vian (gabriel-vian-24) said : 		#6

Dear Olivier,

Thank you for your patience and helpful explanation.

Best regards,
Gabriel V.

Revision history for this message
Gabriel Vian (gabriel-vian-24) said : 		#8

Thanks Olivier Mattelaer, that solved my question.