regarding add process in madgraph

Dear Disha,

Note that the default value of the run_card depends of the process.
In your case the default value in that card are the same for "p p > Z " and "p p > Z J"
but is different from the case when you ask both.

The reason is actually simple:
The process:
p p > Z j
is actually include whithin
p p > Z
Since the cross-section computed are INCLUSIVE with additional radiation.
So asking both simultaneously lead to double counting.

When you ask both simultaneously, what makes sense to compute is
p p > Z EXCLUSIVELY
and
p p > Z j INCLUSIVELY
This method of computation is called matching/merging.

By default when you ask for both process together.
We do activate all the flags for MLM by default.
Note that the cross-section is still not physical and still has double counting.
You have to run the parton-shower in the associated mode (and choose a merging scale).

You will see that the parton-shower is NOT unitary anymore in that case. And that the cross-section will decrease out of this process. This is simply due to the final part of the conversion of
the p p > Z inclusive to a p p > Z exclusive sample.
(you will also have smaller sample size)

See: https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/IntroMatching <https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/IntroMatching>

Cheers,

Olivier

> On 6 Apr 2020, at 14:59, Disha Bhatia <email address hidden> wrote:
>
> New question #689739 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/689739
>
> Hi,
>
> I hope you guys are fine during this pandemic time.
> Sorry for troubling you in these tough times.
>
> I have a very elementary question regarding add process command.
>
> I was under the impression that add process commands simply combines two processes with different final states
> and the total cross-section is the sum of the cross-sections of the individual channels combined.
>
> But strangely, I am not getting so.
>
> I generate p p >z , (z > ve ve~) with cs = 2804 pb
> I then separately generated p p >z j, (z > ve ve~) with cs = 770.18 pb
>
> However, when I generated these two processes using add process command i.e
> generate p p >z , (z > ve ve~)
> add process p p >z j, (z > ve ve~)
> I got the total cross-section equals to 3356 pb which is less than the sum of two.
>
> I am not understanding how come for two different processes whose interference is zero why is this happening.
>
> Thanks,
> Disha
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.