MadGraph5_aMC@NLO

p p > Z l+ l- , Z > l+ l-

Asked by Andre Sznajder

Hi Olivier,
Meng Lu Asked a question 696360 on 2021-04-02 entitled "identical particles affect cross section"
I'm worried about your comments above because we are generating a similar process as background to VBF Higgs up to 3 jets:
 p p > Z l+ l- @1
 p p > Z l+ l- j @2
 p p > Z l+ l- j j @3
 p p > Z l+ l- j j j @4

The only difference is that we decay the Z using Madspin and we are merging different jet multiplicities.
Are we also affected by the symmetry factor in this case ?
Could you elaborate a bit more on why symmetry factor can not be set in a consistent way ?
I remember we already used a similar syntax to generate this process for Higgs production using HEFT ( p p > Z l+ l- , Z > l+ l- ).
Is this also problematic ?
Thanks,
Andre

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

The issue is discussed here:
https://bugs.launchpad.net/mg5amcnlo/+bug/1367374

Revision history for this message
Andre Sznajder (andre-sznajder) said : 		#2

I still have two questions concerning this issue.

1) If I use Madspin to decay Z > l+ l- , am I also affected by this ambiguity ?

2) The correct way to generate my backgrround process would be:
 p p > l+ l- l+ l- @1
 p p > l+ l- l+ l- j @2
 p p > l+ l- l+ l- j j @3
 p p > l+ l- l+ l- j j j @4

As we require at the analysis level that one pair has a mass close to the Z pole, is there a way to require ONE intermediate s-channel Z without breaking gauge invariance and getting non-physical results ?

3) Can Madgraph handle a ME process with 7 fermions in the final state like the above process ?

Thanks,
Andre

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

> 1) If I use Madspin to decay Z > l+ l- , am I also affected by this
> ambiguity ?

Should be.

> 2) The correct way to generate my backgrround process would be:
> p p > l+ l- l+ l- @1
> p p > l+ l- l+ l- j @2
> p p > l+ l- l+ l- j j @3
> p p > l+ l- l+ l- j j j @4

One point to test is to see if you can decompose the above in a sum of contribution.
If you consider only your @1, would this be close enough of

generate p p > z z, z > l+ l- # double resonances
add process p p > z l+ l- $ Z, Z > l+ l- # Exactly one resonances
add process p p > l+ l- l+ l- $ Z # exactly 0 resonance

If it does this, then depending of your analysis you might be able to not include the three process in your analysis and gain in efficiency.

> 3) Can Madgraph handle a ME process with 7 fermions in the final state like the above process ?

Not on a laptop but this should be fine yes.

Cheers,

Olivier

> On 22 Apr 2021, at 17:01, Andre Sznajder <email address hidden> wrote:
>
> Question #696694 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/696694
>
> Status: Answered => Open
>
> Andre Sznajder is still having a problem:
> I still have two questions concerning this issue.
>
> 1) If I use Madspin to decay Z > l+ l- , am I also affected by this
> ambiguity ?
>
> 2) The correct way to generate my backgrround process would be:
> p p > l+ l- l+ l- @1
> p p > l+ l- l+ l- j @2
> p p > l+ l- l+ l- j j @3
> p p > l+ l- l+ l- j j j @4
>
> As we require at the analysis level that one pair has a mass close to
> the Z pole, is there a way to require ONE intermediate s-channel Z
> without breaking gauge invariance and getting non-physical results ?
>
> 3) Can Madgraph handle a ME process with 7 fermions in the final state like the above process ?
>
> Thanks,
> Andre
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Andre Sznajder (andre-sznajder) said : 		#4

Looking at your answer for question https://answers.launchpad.net/mg5amcnlo/+question/696742 , I think we are safe with this process. We impose an analysis level cut 118<m4l<130GeV , so we never have a double resonating Z and there should be no ambiguity , correct ?

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

This is my guess but it would be better to perform the check I mention above to check if this is indeed the case or not.

Cheers,

Olivier

> On 26 Apr 2021, at 13:30, Andre Sznajder <email address hidden> wrote:
>
> Question #696694 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/696694
>
> Status: Answered => Open
>
> Andre Sznajder is still having a problem:
> Looking at your answer for question
> https://answers.launchpad.net/mg5amcnlo/+question/696742 , I think we
> are safe with this process. We impose an analysis level cut
> 118<m4l<130GeV , so we never have a double resonating Z and there
> should be no ambiguity , correct ?
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Andre Sznajder (andre-sznajder) said : 		#6

If we generate the processes ( p p > l+ l- l+ l- ) and ( p p > Z l+ l- , Z > l+ l- ) and compare the remaining cross sections after the 118<m4l<130GeV cut , would you consider that enough to validate the claim that we are not being affected by the ambiguity in the symmetry factor ?
Cheers,
Andre

Revision history for this message
Andre Sznajder (andre-sznajder) said : 		#7

Hi Olivier,
Indeed , we made the a test along the lines you suggested above and even with the 118<m4l<130GeV cut we still get a reasonable contribution from diagrams with ZERO resonant Z's ( p p > l+ l- l+ l- $ Z ) . So we can't use the process ( p p > Z l+ l- , Z > l+ l- ) anyway ...
Thanks,
Andre

Revision history for this message
Andre Sznajder (andre-sznajder) said : 		#8

Thanks Olivier Mattelaer, that solved my question.