Running b yukawa

Asked by aravind on 2018-01-21

Hello,

     I was studying the charged higgs production cross section at 13 TeV LHC using type 2 two higgs doublet model (http://feynrules.irmp.ucl.ac.be/attachment/wiki/2HDM/2HDMtII_NLO.tar.gz) of mass 500 GeV and tan beta of 30 (in four flavor scheme of PDF).
     I wanted to know if madgraph uses correct running b quark yukawa (as the cross section is highly sensitive to this) or this needs to be entered by hand (in the model parameters file) ?
     Also I am getting NLO cross section to be smaller than LO cross section, is this fine ? Also the cross sections I am getting are significantly different for 4 flavour scheme and 5 flavor scheme of PDF (after making incoming b quark massless) and disagrees significantly with the higgs cross section working group predictions.

Regards
Aravind.

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
2018-01-22
Last query:
2018-01-22
Last reply:
2018-01-21
aravind (aravindhv10) said : #1

Sorry, forgot to mention the actual process, it is:

define top = t t~
define bot = b b~
define hig = h+ h-
generate p p > top bot hig

Hi,

I do not know the model, the more secure solution would be to contact the author of the model.

> Also I am getting NLO cross section to be smaller than LO cross-section, is this fine?

This is possible, especially if you do not use the same dynamical scale choice at LO and NLO, the typical error for the scale at LO being 50%, it is expected that some NLO cross-section can reduce the cross-section (within the uncertainty).
Note that default cuts at LO/NLO (and scheme) are not the same, so in MG5aMC if you want to compute the K-factor, you should not use the LO generation of MG5aMC, but ask the NLO output to compute the LO cross-section.

 > Also the cross sections I am getting are significantly different for 4 flavour scheme and 5 flavour scheme of PDF (after making incoming b quark massless)

This is also something possible. I would recommend you to read some paper on 4/5 flavour comparison and when to use which one(for example:http://inspirehep.net/record/1382155).
They do not have to agree on the full spectrum.

> disagrees significantly with the Higgs cross section working group predictions.

Which means that you are not computing the same number as they do (not the same cuts or the same precision, ...)

Cheers,

Olivier

aravind (aravindhv10) said : #3

Hello,

     Thanks for a quick reply especially on a Sunday ! I will check the reference you mentioned and get back. I also wanted to know how this is treated in default standard model of madgraph. Consider a process for example b b~ > h , at tree level or at NLO, does madgraph take care of the running b mass when it calculates the cross section or is it fixed ?

Thanks and Regards
Aravind

Hi,

Our default standard model can not do b b~ > h at NLO.
Indeed our model has the yukawa coupling equal to the mass (This is due to the re-normalization scheme). Therefore if you try to compute a 5 flavor computation with our model,
the b-quark does not couple to the Higgs.

At LO, the two parameter are disjoint and you can do it but they are no running associated to either the mass or the coupling.

Natively (i.e. if not defined as part of the model) the only running is the one related to as.

Cheers,

Olivier

> On 21 Jan 2018, at 15:08, aravind <email address hidden> wrote:
>
> Question #663362 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/663362
>
> aravind posted a new comment:
> Hello,
>
> Thanks for a quick reply especially on a Sunday ! I will check the
> reference you mentioned and get back. I also wanted to know how this is
> treated in default standard model of madgraph. Consider a process for
> example b b~ > h , at tree level or at NLO, does madgraph take care of
> the running b mass when it calculates the cross section or is it fixed ?
>
> Thanks and Regards
> Aravind
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

aravind (aravindhv10) said : #5

Thanks Olivier Mattelaer, that solved my question.