Check MLM matching

I attach the plot for your convenience.

Cheers,
Marco

On Sat, Jul 8, 2017 at 11:53 AM, marco santoni <
<email address hidden>> wrote:

> New question #647109 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/647109
>
> Dear authors,
>
> I am generating a mssm process
>
> generate p p > x1+ x1- @0
> add process p p > x1+ x1- j @1
>
> with masses 250 GeV for the chargino (200 GeV for the LSP).
>
> Since I am requiring an ISR jet I apply an MLM matching in the context
> MadGraph5 + Pythia6.
> For example, I choose the values xqcut = 50 , QCUT = 60
>
> Checking the differential jet rate distribution (DJR1), the transition
> appears smooth between the 0->1 jet-samples, but the 0-sample line does not
> stop (it continues after the value ~2).
> As a consequence, the sum-of-contributions distro has a step at the
> transition.
>
> My questions are:
>
> 1) Am I double counting or doing something wrong?
> In other words, I expected the 0-lepton line to go to 0 immediately after
> the transition.
>
> 2) Do you suggest the following?: the more compressed the spectrum
> (Mchargino-MLSP smaller) the less the xqcut value.
>
> Thanks in advance for your help.
>
> Cheers,
> Marco
>
>
>
>
>
>
> --
> You received this question notification because you asked the question.
>

Hi,

I do not have the plot.

> 1) Am I double counting or doing something wrong?
> In other words, I expected the 0-lepton line to go to 0 immediately after the transition.

This "0j" contribution is likely to be a "1j" contribution tagged as 0j since those diagram do not have any particle that can be consider as a QCD radiation.
If you generate the two sample separately and indicate to Pythia the highest multiplicity sample to consider you will see that indeed
The 0j contribution of the 0j sample stops at the merging value.
And that the 1j sample will have two contribution on the plot, one tagged as 0j (and present both before and after the merging scale) and one tagged as 1j
Which should start exactly at the merging scale.

If you look at the Feynman diagram for g u initial state, you can see the diagram #4 which is one of those type of non QCD contribution.
We call that effect, diagram opening. This being said, your DJR should be smooth anyway. I would personally increase the gap between xqcut and qcut.

> 2) Do you suggest the following?: the more compressed the spectrum (Mchargino-MLSP smaller) the less the xqcut value.\\\

No I would not suggest that, since I would expect that the ISR should be insensitive to the compression of the spectrum.

Cheers,

OLIVIER

> On 8 Jul 2017, at 04:23, marco santoni <email address hidden> wrote:
>
> New question #647109 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/647109
>
> Dear authors,
>
> I am generating a mssm process
>
> generate p p > x1+ x1- @0
> add process p p > x1+ x1- j @1
>
> with masses 250 GeV for the chargino (200 GeV for the LSP).
>
> Since I am requiring an ISR jet I apply an MLM matching in the context MadGraph5 + Pythia6.
> For example, I choose the values xqcut = 50 , QCUT = 60
>
> Checking the differential jet rate distribution (DJR1), the transition appears smooth between the 0->1 jet-samples, but the 0-sample line does not stop (it continues after the value ~2).
> As a consequence, the sum-of-contributions distro has a step at the transition.
>
> My questions are:
>
> 1) Am I double counting or doing something wrong?
> In other words, I expected the 0-lepton line to go to 0 immediately after the transition.
>
> 2) Do you suggest the following?: the more compressed the spectrum (Mchargino-MLSP smaller) the less the xqcut value.
>
> Thanks in advance for your help.
>
> Cheers,
> Marco
>
>
>
>
>
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

 Dear Olivier,

thank you very much for the prompt and precise answers.
They have been really helpful.

>This "0j" contribution is likely to be a "1j" contribution tagged as 0j
since those diagram do not have any particle that can be consider as a QCD
radiation.

This is clear. At this regard, there is a way to modify the legend? Also
because when I generate processes involving SUSY particles with larger
masses the distributions at the merging scale are below the legend.

I have read xquct between 1/6 and 1/3 the hard scale and QCUT 20% larger,
but a QCUT 50% larger (or more) seems to work maybe better. (I am not using
the "Shower kT" scheme) I try to put the plots for DJR1 and DJR2 and for
the same process as before, but till 2 extra jets and with xqcut=50 and
QCUT=80. Hence, they are fine!?

>If you look at the Feynman diagram for g u initial state, you can see the
diagram #4 which is one of those type of non QCD contribution. We call that
effect, diagram opening.

That's interesting. Those kinds of processes for the generation of
electroweakinos weren't at the top of my mind. Of course is not QCD.

>I would expect that the ISR should be insensitive to the compression of
the spectrum.

Yes, I thought the same. It seemed to appear a strange effect varying the
mass of the neutralino for the process
generate p p > b1 b1~, (b1 > n1 b), (b1~ > n1 b~) @0
add process p p > b1 b1~ j, (b1 > n1 b), (b1~ > n1 b~) @1
...
But I need to check better at the transition for masses of sbottom 1 ~ 1.5
TeV. At this regard, I cannot see very well since the legend is a problem.

As the last question: for this process, (I maintain maxjetflavor 4
(defining j without bottoms)), how many additional QCD radiations would you
add in the generation? The average jet multiplicity I observe at
the Delphes level is around 4.5-5 (independently on the tagging).

Thanks again.

Cheers,
Marco

​​

On Sat, Jul 8, 2017 at 4:39 PM, Olivier Mattelaer <
<email address hidden>> wrote:

> Your question #647109 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/647109
>
> Status: Open => Answered
>
> Olivier Mattelaer proposed the following answer:
> Hi,
>
> I do not have the plot.
>
> > 1) Am I double counting or doing something wrong?
> > In other words, I expected the 0-lepton line to go to 0 immediately
> after the transition.
>
> This "0j" contribution is likely to be a "1j" contribution tagged as 0j
> since those diagram do not have any particle that can be consider as a QCD
> radiation.
> If you generate the two sample separately and indicate to Pythia the
> highest multiplicity sample to consider you will see that indeed
> The 0j contribution of the 0j sample stops at the merging value.
> And that the 1j sample will have two contribution on the plot, one tagged
> as 0j (and present both before and after the merging scale) and one tagged
> as 1j
> Which should start exactly at the merging scale.
>
> If you look at the Feynman diagram for g u initial state, you can see the
> diagram #4 which is one of those type of non QCD contribution.
> We call that effect, diagram opening. This being said, your DJR should be
> smooth anyway. I would personally increase the gap between xqcut and qcut.
>
> > 2) Do you suggest the following?: the more compressed the spectrum
> (Mchargino-MLSP smaller) the less the xqcut value.\\\
>
> No I would not suggest that, since I would expect that the ISR should be
> insensitive to the compression of the spectrum.
>
> Cheers,
>
> OLIVIER
>
> > On 8 Jul 2017, at 04:23, marco santoni <question647109@answers.
> launchpad.net> wrote:
> >
> > New question #647109 on MadGraph5_aMC@NLO:
> > https://answers.launchpad.net/mg5amcnlo/+question/647109
> >
> > Dear authors,
> >
> > I am generating a mssm process
> >
> > generate p p > x1+ x1- @0
> > add process p p > x1+ x1- j @1
> >
> > with masses 250 GeV for the chargino (200 GeV for the LSP).
> >
> > Since I am requiring an ISR jet I apply an MLM matching in the context
> MadGraph5 + Pythia6.
> > For example, I choose the values xqcut = 50 , QCUT = 60
> >
> > Checking the differential jet rate distribution (DJR1), the transition
> appears smooth between the 0->1 jet-samples, but the 0-sample line does not
> stop (it continues after the value ~2).
> > As a consequence, the sum-of-contributions distro has a step at the
> transition.
> >
> > My questions are:
> >
> > 1) Am I double counting or doing something wrong?
> > In other words, I expected the 0-lepton line to go to 0 immediately
> after the transition.
> >
> > 2) Do you suggest the following?: the more compressed the spectrum
> (Mchargino-MLSP smaller) the less the xqcut value.
> >
> > Thanks in advance for your help.
> >
> > Cheers,
> > Marco
> >
> >
> >
> >
> >
> >
> > --
> > You received this question notification because you are an answer
> > contact for MadGraph5_aMC@NLO.
>
> --
> If this answers your question, please go to the following page to let us
> know that it is solved:
> https://answers.launchpad.net/mg5amcnlo/+question/647109/+
> confirm?answer_id=1
>
> If you still need help, you can reply to this email or go to the
> following page to enter your feedback:
> https://answers.launchpad.net/mg5amcnlo/+question/647109
>
> You received this question notification because you asked the question.
>

Thanks Olivier Mattelaer, that solved my question.

Hi Marco,

I guess that I understood the problem now,
You have kept(and/or set) 'use_syst' on T.
This prevent the Parton-shower to veto the events, since in that case, it should be used with syscalc such that such tools can decide to veto the event or not afterwards.
In that context, all the automatic plot are actually wrong. (This is why use_syst was on OFF by default for matching and merging).

Sorry to not have catch the problem now, but this is mainly because i have forgetten about such problem since we do not support those tools since rouhgly a year now.

Cheers,

Olivier