Calculate cross-section corresponding to aQGC reweight parameters

Asked by Ramkrishna on 2017-08-05

Dear experts,

When we generate LHE file using reweight card then to get the cross-section corresponding to each parameter, I am doing following:

Cross-section for aQGC point k

= Average of (wgt[k]/wgt[0]) * (Total cross-section given in gridpack)

Is this correct method to get the cross-section??

I calculated the SM cross-section for p p -> W+ W- JJ > l+ vl QQ JJ. For this process i got cross-section 0.91pb

When I calculated the reweighted SM cross-section from a aQGC point it is giving me value 0.228pb.

These two values should be consistent with each other but I am getting difference of factor of ~4.

Let me know if you need more information.

with regards,
Ram

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
2017-08-09
Last reply:
2017-08-09

Dear Ram,

> Cross-section for aQGC point k
>
> = Average of (wgt[k]/wgt[0]) * (Total cross-section given in gridpack)
>
> Is this correct method to get the cross-section??

No this should be either
> Average of (wgt[k])
or
> Sum of (wgt[k])

Depending of your choice of normalisation in the run_card. Those weights have the same normalisation as the central weight.

Cheers,

Olivier

> On 5 Aug 2017, at 13:33, Ramkrishna <email address hidden> wrote:
>
> New question #655340 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/655340
>
> Dear experts,
>
> When we generate LHE file using reweight card then to get the cross-section corresponding to each parameter, I am doing following:
>
> Cross-section for aQGC point k
>
> = Average of (wgt[k]/wgt[0]) * (Total cross-section given in gridpack)
>
> Is this correct method to get the cross-section??
>
> I calculated the SM cross-section for p p -> W+ W- JJ > l+ vl QQ JJ. For this process i got cross-section 0.91pb
>
> When I calculated the reweighted SM cross-section from a aQGC point it is giving me value 0.228pb.
>
> These two values should be consistent with each other but I am getting difference of factor of ~4.
>
> Let me know if you need more information.
>
> with regards,
> Ram
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Dear Oliver,

Thanks for help.

The run card that I used had no event normalization option (http://bit.ly/2v95L5f)

But, I calculated both numbers sum and average and realized that in my case it should be average. So, the reweighted cross-section that I am getting is 0.000872792 (unit???) and the SM cross-section that I got from madgraph is 0.91pb.

Here I am unable to get the unit of the reweighted SM cross-section?

Please help me with this.

Thanks & regards,
Ram

PS: The value that I got from Sum of (Wgt[k]) is 1431.53. So, it does not seems like cross-section.

Your unit should be in pb.

Did you try to switch off the polarisation? If you do not care about the longitudinal/transverse state of your particles, you can switch that off for the re-weighitng.
It would expect that your EFT theory enhances the longitudinal component which is not the case for the SM leading to some potential convergence issue.

Cheers,

Olivier

> On 6 Aug 2017, at 14:33, Ramkrishna <email address hidden> wrote:
>
> Question #655340 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/655340
>
> Status: Answered => Open
>
> Ramkrishna is still having a problem:
> Dear Oliver,
>
> Thanks for help.
>
> The run card that I used had no event normalization option
> (http://bit.ly/2v95L5f)
>
> But, I calculated both numbers sum and average and realized that in my
> case it should be average. So, the reweighted cross-section that I am
> getting is 0.000872792 (unit???) and the SM cross-section that I got
> from madgraph is 0.91pb.
>
> Here I am unable to get the unit of the reweighted SM cross-section?
>
> Please help me with this.
>
> Thanks & regards,
> Ram
>
> PS: The value that I got from Sum of (Wgt[k]) is 1431.53. So, it does
> not seems like cross-section.
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Dear Olivier,

I know how to get a particular polarization but How I can switch off the polarization?

So, As you are saying that EFT enhances the longitudinal component So, it's not exactly same as SM. I will try to generate only SM from EFT, then it should give me similar cross-section as I am getting here from reweight.

I will do this and let you know what I found.

with regards,
Ram

Your unit should be in pb.

Did you try to switch off the polarisation? If you do not care about the longitudinal/transverse state of your particles, you can switch that off for the re-weighitng.
It would expect that your EFT theory enhances the longitudinal component which is not the case for the SM leading to some potential convergence issue.

Cheers,

Olivier

> On 6 Aug 2017, at 14:33, Ramkrishna <email address hidden> wrote:
>
> Question #655340 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/655340
>
> Status: Answered => Open
>
> Ramkrishna is still having a problem:
> Dear Oliver,
>
> Thanks for help.
>
> The run card that I used had no event normalization option
> (http://bit.ly/2v95L5f)
>
> But, I calculated both numbers sum and average and realized that in my
> case it should be average. So, the reweighted cross-section that I am
> getting is 0.000872792 (unit???) and the SM cross-section that I got
> from madgraph is 0.91pb.
>
> Here I am unable to get the unit of the reweighted SM cross-section?
>
> Please help me with this.
>
> Thanks & regards,
> Ram
>
> PS: The value that I got from Sum of (Wgt[k]) is 1431.53. So, it does
> not seems like cross-section.
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

For the helicity,
You can set "change helicity False"
In the reweight card.
In that case the re-weighting will ignore the helicity information (i.e. you can not use that information anymore)

Cheers,

Olivier

> On 6 Aug 2017, at 16:38, Ramkrishna <email address hidden> wrote:
>
> Question #655340 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/655340
>
> Status: Answered => Open
>
> Ramkrishna is still having a problem:
> Dear Olivier,
>
> I know how to get a particular polarization but How I can switch off the
> polarization?
>
> So, As you are saying that EFT enhances the longitudinal component So,
> it's not exactly same as SM. I will try to generate only SM from EFT,
> then it should give me similar cross-section as I am getting here from
> reweight.
>
> I will do this and let you know what I found.
>
> with regards,
> Ram
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Dear Olivier,

Thanks for help.

I generated SM only (without any reweight card) from EFT by putting all the aQGC parameters to zero. So, now I am summarizing things :

1. SM Cross-Section = 0.911 pb

2. SM cross-section from EFT = 0.2062 pb

3. Reweighted SM Cross-section within EFT calculated with [ Average of (wgt[k]/wgt[0]) * (Total cross-section given in gridpack) ] = 0.222191

4. Reweighted SM Cross-section within EFT calculated with [ Average of (wgt[k]) ] = 0.00087

==> So, by looking at 2nd, 3rd and 4th point it seems that the reweighted cross-section should be calculated using formula [Average of (wgt[k]/wgt[0]) * (Total cross-section given in gridpack) ]!!!

==> I already added the line "change helicity False" in the reweighted cards [1], then why there is the difference between the two SM's (1st and the 2nd point above)? Is it because helicity is on in the SM?? How can I switch off helicity in the SM generation where I don't have reweight card?

==> I generated SM and SM within EFT from the script [2] and [3] respectively. And all the aQGC points (i.e. grid pack) using reweight was generated using the cards [4].

Let me know if you need more information.

with regards,
Ram

[1] https://github.com/cms-sw/genproductions/blob/master/bin/MadGraph5_aMCatNLO/cards/production/13TeV/VBS/VVjj_semileptonic/aQGC/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10_reweight_card.dat

[2] http://rasharma.web.cern.ch/rasharma/MC_Study_VBS/temp/WPlepWMhadJJ_SM_LO_EWK_mjj100_pTj10_proc_card.txt

[3] http://rasharma.web.cern.ch/rasharma/MC_Study_VBS/temp/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10_proc_card.txt

[4] https://github.com/cms-sw/genproductions/tree/master/bin/MadGraph5_aMCatNLO/cards/production/13TeV/VBS/VVjj_semileptonic/aQGC/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10

Hi,

> ==> So, by looking at 2nd, 3rd and 4th point it seems that the
> reweighted cross-section should be calculated using formula [Average of
> (wgt[k]/wgt[0]) * (Total cross-section given in gridpack) ]!!!

I actually do not understand how your method 3 and 4 do not give the same answer.
Let's me explains in details. Such that you can check which assumptions is wrong in your case

1) the weights wgt[0] correspond to the central weight
Therefore wgt[0] is the same for ALL events

2) you use the event_norm=average and therefore
1/N\sum wgt[0] = cross-section

3) since point 1) we have
1/N*N *wgt[0] = cross-section

4) therefore If I look at your equation "3"

1/N \sum[ wgt[k]/wgt[0]) ] * cross-section
= 1/N \sum[ wgt[k] ] * cross-section/wgt[0] # since wgt[0] is constant
= 1/N \sum[ wgt[k] ] # since wgt[0] = cross-section
This last equation is nothing else than your equation "4"

Consequently, using "3" or "4" should be equivalent.
Since this is not the case, then you are not using "event_norm=average" did you? (Or your generation is not un-weighted to start with)

> ==> I already added the line "change helicity False" in the reweighted
> cards [1], then why there is the difference between the two SM's (1st
> and the 2nd point above)? Is it because helicity is on in the SM?? How
> can I switch off helicity in the SM generation where I don't have
> reweight card?

At generation time, it does not make any difference to have this turn on or not. So we do not have any switch to turn it off.

One other point which can explain difference between your generation is the choice of the renormalisation and factorisation scale.
If you use the default of MG5aMC, then this is process and FD dependent. Since the reweighing does not change the scale choice, you can have different
Results up to those uncertainty. To avoid that you need to choose another scale.

Cheers,

Olivier

> On 9 Aug 2017, at 18:09, Ramkrishna <email address hidden> wrote:
>
> Question #655340 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/655340
>
> Status: Answered => Open
>
> Ramkrishna is still having a problem:
> Dear Olivier,
>
> Thanks for help.
>
> I generated SM only (without any reweight card) from EFT by putting all
> the aQGC parameters to zero. So, now I am summarizing things :
>
> 1. SM Cross-Section = 0.911 pb
>
> 2. SM cross-section from EFT = 0.2062 pb
>
> 3. Reweighted SM Cross-section within EFT calculated with [ Average of
> (wgt[k]/wgt[0]) * (Total cross-section given in gridpack) ] = 0.222191
>
> 4. Reweighted SM Cross-section within EFT calculated with [ Average of
> (wgt[k]) ] = 0.00087
>
> ==> So, by looking at 2nd, 3rd and 4th point it seems that the
> reweighted cross-section should be calculated using formula [Average of
> (wgt[k]/wgt[0]) * (Total cross-section given in gridpack) ]!!!
>
> ==> I already added the line "change helicity False" in the reweighted
> cards [1], then why there is the difference between the two SM's (1st
> and the 2nd point above)? Is it because helicity is on in the SM?? How
> can I switch off helicity in the SM generation where I don't have
> reweight card?
>
> ==> I generated SM and SM within EFT from the script [2] and [3]
> respectively. And all the aQGC points (i.e. grid pack) using reweight
> was generated using the cards [4].
>
> Let me know if you need more information.
>
> with regards,
> Ram
>
>
> [1] https://github.com/cms-sw/genproductions/blob/master/bin/MadGraph5_aMCatNLO/cards/production/13TeV/VBS/VVjj_semileptonic/aQGC/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10_reweight_card.dat
>
> [2]
> http://rasharma.web.cern.ch/rasharma/MC_Study_VBS/temp/WPlepWMhadJJ_SM_LO_EWK_mjj100_pTj10_proc_card.txt
>
> [3]
> http://rasharma.web.cern.ch/rasharma/MC_Study_VBS/temp/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10_proc_card.txt
>
> [4] https://github.com/cms-
> sw/genproductions/tree/master/bin/MadGraph5_aMCatNLO/cards/production/13TeV/VBS/VVjj_semileptonic/aQGC/aQGC_WPlepWMhadJJ_EWK_LO_NPle1_mjj100pt10
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Can you help with this problem?

Provide an answer of your own, or ask Ramkrishna for more information if necessary.

To post a message you must log in.