MadGraph5_aMC@NLO

Using same gridpack for multiple merging schemes

Asked by Nicolas Gutierrez

Dear all,

We are currently trying to find the ideal merging scheme
for the production of Z+jets and W+jets.

Those samples are sliced using the 'ihtmin' and 'ihtmax' parameters,
for a total of 8 slices, and each has a corresponding gridpack.
Generating these gridpacks is quite a task, which we would like to
avoid repeating unless it is absolutely necessary.

We use Py8 for showering.
The showering is done running Py8 independently from MG5.
Can we shower multiple times these gridpacks,
using each time a different merging scheme?
Please note that we are currently not using SysCalc or other
event reweighting strategy for the evaluation of uncertainties.

Thanks,
Nicolas

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

yes you can change the matching scale used by pythia8 (as long as it remains well above the xqcut value)

Cheers,

Olivier

Revision history for this message
Simon Berlendis (simon-berlendis) said : 		#2

 Hi Olivier,

 What's about the merging scheme ?
 Can we use a different the merging scheme with the same gridpack ?

 Cheers, Simon.
 (working with Nicolas)

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

That's the same scale which should be named JetMatching:qCut

Cheers,

Olivier

Revision history for this message
Josh McFayden (mcfayden) said : 		#4

I think Olivier may need a little info, so to elaborate a bit:

The question here is whether it is possible to use gridpacks with ktdurham=30 and dparameter=0.2 designed for CKKW-L merging with MLM matching too?
I'm 99% sure the answer is no because we would need to set xqcut and icckw=1 etc. to use MLM, but we just want to be absolutely certain.
But I also remember seeing that one can get Pythia8 to apply the cuts necessary for MLM, is there any way that we can could apply harder cuts to the ME in Py8 and still use these gridpacks?
Again, I expect it is not possible (or at least not very safe) but it would be good to get confirmation.

Best,

Josh.

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

Hi,

Sorry Simon/Nicolas, I missread your question.

Changing from CKKWL to MLM is something extremely unsafe. That I would not recommend.
In some scenario this is theoritically possible, but since the partonic level cut are not the same at parton level, you will probably need to use extremely soft cut in order to have something sensitive. In turn the efficiency after matching/merging will then significantly drop which is inefficient.

Additionally for MLM, the afas/PDF reweighting is done by MG, while it is done by PY8 for CKKWL.
I'm pretty sure that PY8 can perform a MLM matching merging on sample who does not have such reweighitng but no idea how to do it.

Actually we were considering something like that for the new MG/PY8 interface (under development) but we finally discard the idea due to those associated difficulty.

Cheers,

Olivier

Can you help with this problem?

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

To post a message you must log in.