MadGraph5_aMC@NLO

Bug #1865848
Comment #8

Comment 8 for bug 1865848

Revision history for this message

Olivier Mattelaer (olivier-mattelaer) wrote on 2020-03-03: Re: [Bug 1865848] I saw in DJR plots, the plot goes to zero before log(DJR) = 3.0 value. But in my DJR1 plot for e- e+ > t t~ + jets , the plot continues after 3.0. Is it correct?

Hi,

> Anyway I still could
> not change the range of x axis more than 3.0 in DJR plot to see the tail
> of the plot.

That is not a question for me, but for the MadAnalysis5 team.

> The main problem is that I expected (for example in 20_30.png) the blue
> line (0-jet sample) go to zero around the peak of the red line (1-jet
> sample) to distinguish between PS and ME. But it did not happen. Do you
> have any idea about it?

Most tools are not able to plot the TRUE DJR plot. They are typically plotting proxies...
Which makes it difficult for validation.
This is why I liked to validate MLM with pythia6 and madanalysis4 since that plot was the True DJR.
PY8 authors themself claims that they can not plot the True DJR for pythia8 ( i do not think this is true but this is what they claim)

Cheers,

Olivier

> On 3 Mar 2020, at 11:46, Reza Jafari <email address hidden> wrote:
>
> Dear Olivier,
>
> Thanks a lot for your useful response.
>
> Actually I prepared DJR plots for very different values of xqcut and
> QCUT from 10 to 60 for both. For example see the new attachment for
> xqcut = 30 and QCUT = 60 (30_60.png). I will extend QCUT value to 100
> GeV as you suggested and let you know the result. Anyway I still could
> not change the range of x axis more than 3.0 in DJR plot to see the tail
> of the plot.
>
> The main problem is that I expected (for example in 20_30.png) the blue
> line (0-jet sample) go to zero around the peak of the red line (1-jet
> sample) to distinguish between PS and ME. But it did not happen. Do you
> have any idea about it?
>
> I would greatly appreciated for your time and advises.
>
> Best Regards,
>
> Reza
>
>
>
>
> ** Attachment added: "30_60.png"
> https://bugs.launchpad.net/mg5amcnlo/+bug/1865848/+attachment/5333020/+files/30_60.png
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1865848
>
> Title:
> I saw in DJR plots, the plot goes to zero before log(DJR) = 3.0 value.
> But in my DJR1 plot for e- e+ > t t~ + jets , the plot continues after
> 3.0. Is it correct?
>
> Status in MadGraph5_aMC@NLO:
> New
>
> Bug description:
> Dear experts,
>
> I am working on a project on top quark at the electron-positron colliders, e-e+ -> ttbar+(jet) where I have to perform merging.
> I performed the MLM matching and for sanity check I prepared the DJR plots. I would be very much grateful if you help me with the following explanation and give advice.
>
> - I plotted the DJR1 figure for e- e+ > t t~ + jets process. In our DJR1 plot (0 jet sample -> 1 jet sample), the range of DJR parameter is not like the others. In my DJR2, DJR3 (and also DJR1 in other processes that I saw in some presentations), plots go to zero before log(DJR) = 3.0. But in my DJR1 plot, it continues after 3.0 value. I am not sure whether this plot is correct or not?
> If log(DJR1) can take values larger than 3.0, how can we change the range of the parameter in the plot?
>
> - In order to compare the results of matching (just for 0-jet and 1-jet samples), I chose different values for xqcut (in Madgraph run card) and qCut (in pythia card). I prepared DJR plot for below values:
> xqcut = 10 , qCut = 10
> xqcut = 20 , qCut = 20
> xqcut = 20 , qCut = 30
> I don't have any idea which values for xqcut and qCut parameters are the best? And which plot is suitable for an acceptable matching? I think all plots are smooth with little differences.
>
> I would greatly appreciate any assistance or advice you could offer
> me.
>
> Thank you very much for your time.
> Best regards,
>
> Reza
>
>
> PS.
> Some more details of what we have done are as follows.
>
> MadGraph:
> C.M. energy = 3 TeV
> generate e- e+ > t t~ @0
> add process e- e+ > t t~ j @1
> add process e- e+ > t t~ j j @2
> add process e- e+ > t t~ j j j @3
> ickkw = 1, xqcut = 10 or 20 or ... , qCut = 10 or 20 or 30, ...
>
> MadAnalysis:
> bin/ma5 -H
> ma5> install fastjet
> ma5> import <path-to-the-event-file>/pythia_events.hep.gz as ttjets
> ma5> set main.merging.check = true
> ma5> set main.merging.njets = 3 (for our purpose = 1)
> ma5> submit output
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1865848/+subscriptions

Hi,

>  Anyway I still could
> not change the range of x axis more than 3.0 in DJR plot to see the tail
> of the plot.

That is not a question for me, but for the MadAnalysis5 team.

Cheers,

Olivier

> On 3 Mar 2020, at 11:46, Reza Jafari <1865848@bugs.launchpad.net> wrote:
> 
> Dear Olivier,
> 
> Thanks a lot for your useful response.
> 
> Actually I prepared DJR plots for very different values of xqcut and
> QCUT from 10 to 60 for both. For example see the new attachment for
> xqcut = 30 and QCUT = 60 (30_60.png). I will extend QCUT value to 100
> GeV as you suggested and let you know the result. Anyway I still could
> not change the range of x axis more than 3.0 in DJR plot to see the tail
> of the plot.
> 
> The main problem is that I expected (for example in 20_30.png) the blue
> line (0-jet sample) go to zero around the peak of the red line (1-jet
> sample) to distinguish between PS and ME. But it did not happen. Do you
> have any idea about it?
> 
> I would greatly appreciated for your time and advises.
> 
> Best Regards,
> 
> Reza
> 
> 
> 
> 
> ** Attachment added: "30_60.png"
>   https://bugs.launchpad.net/mg5amcnlo/+bug/1865848/+attachment/5333020/+files/30_60.png
> 
> -- 
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1865848
> 
> Title:
>  I saw in DJR plots, the plot goes to zero before log(DJR) = 3.0 value.
>  But in my DJR1 plot for e- e+ > t t~ + jets , the plot continues after
>  3.0. Is it correct?
> 
> Status in MadGraph5_aMC@NLO:
>  New
> 
> Bug description:
>  Dear experts,
> 
>  I am working on a project on top quark at the electron-positron colliders, e-e+ -> ttbar+(jet) where I have to perform merging. 
>  I performed the MLM matching and for sanity check I prepared the DJR plots. I would be very much grateful if you help me with the following explanation and give advice.
> 
>  - I plotted the DJR1 figure for e- e+ > t t~ + jets process. In our DJR1 plot (0 jet sample -> 1 jet sample), the range of DJR parameter is not like the others. In my DJR2, DJR3 (and also DJR1 in other processes that I saw in some presentations), plots go to zero before log(DJR) = 3.0. But in my DJR1 plot, it continues after 3.0 value. I am not sure whether this plot is correct or not?
>  If log(DJR1) can take values larger than 3.0, how can we change the range of the parameter in the plot?
> 
>  - In order to compare the results of matching (just for 0-jet and 1-jet samples), I chose different values for xqcut (in Madgraph run card) and qCut (in pythia card). I prepared DJR plot for below values:
>  xqcut = 10 , qCut = 10
>  xqcut = 20 , qCut = 20
>  xqcut = 20 , qCut = 30
>  I don't have any idea which values for xqcut and qCut parameters are the best? And which plot is suitable for an acceptable matching? I think all plots are smooth with little differences.
> 
>  I would greatly appreciate any assistance or advice you could offer
>  me.
> 
>  Thank you very much for your time.
>  Best regards,
> 
>  Reza
> 
> 
>  PS.
>  Some more details of what we have done are as follows.
> 
>  MadGraph:
>  C.M. energy = 3 TeV
>  generate e- e+ > t t~ @0
>  add process e- e+ > t t~ j @1
>  add process e- e+ > t t~ j j @2
>  add process e- e+ > t t~ j j j @3
>  ickkw = 1, xqcut = 10 or 20 or ... , qCut = 10 or 20 or 30, ...
> 
>  MadAnalysis:
>  bin/ma5 -H
>  ma5> install fastjet
>  ma5> import <path-to-the-event-file>/pythia_events.hep.gz as ttjets
>  ma5> set main.merging.check = true
>  ma5> set main.merging.njets = 3 (for our purpose = 1)
>  ma5> submit output
> 
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1865848/+subscriptions