Is it possible to generate pion-proton collisions?

Dear Celso,

No we do not have any hidden option for defining two different PDF on the initial state.
They are no easy solutions: Here is some idea:

- I would suggest to generate your sample with p p > mu+ mu- and the reweight each event, with the ratio of the pdf that you are looking for.
- The second option is actually the same but performing such ratio inside the event generation, by using the Bias module:
https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/LOEventGenerationBias
- The last option is obviously to hack the code to do the same but this is probably the most difficult option

Cheers,

Olivier

Dear Olivier,

Thanks for the suggestions. However, can you be more specific and clarify the issue of reweighting events? In my case I have two sets of PDFs (pion and proton from lhapdf) that I need to use. From your suggestion I should generate p p > mu+ mu- [QCD] (I want NLO) using a proton lhapdf. Then, where and how should I define the ratio of PDFs as an event weight?

Thanks in advance.
Best regards,

Celso

Hi Celso,

You did not mention that it has to be NLO. This actually change things (in an even worse situation).

The bias option is indeed not available at NLO and that was the best option at LO.

Doing a posteriori reweighting is also possible at NLO but it is more complicated.
What you have to do in that case is to generate your sample with the option
store_rwgt_info set on True

Then you need to modify the script systematics.py (madgraph/various/systematics.py )
in order to produce the weight that you want.
This is code is designed to compute LO and NLO theoretical uncertainty which include pdf re-weighting.
In particular you will need to modify the function get_nlo_wgt (line 661).
That file is exectable, so you can run it.
In your case, you can run that file (after your modification) as

python systematics.py —-lhapd_config=XXXXX —-pdf=XXX —-mur=1 --muf=1 -—dyn=1
(Be carefull that mail formatting is likely to make the above line not fully correct)

Cheers,

Olivier

On 29 Mar 2017, at 13:27, Celso <<email address hidden><mailto:<email address hidden>>> wrote:

Question #596745 on MadGraph5_aMC@NLO changed:
https://answers.launchpad.net/mg5amcnlo/+question/596745

   Status: Answered => Open

Celso is still having a problem:
Dear Olivier,

Thanks for the suggestions. However, can you be more specific and
clarify the issue of reweighting events? In my case I have two sets of
PDFs (pion and proton from lhapdf) that I need to use. From your
suggestion I should generate p p > mu+ mu- [QCD] (I want NLO) using a
proton lhapdf. Then, where and how should I define the ratio of PDFs as
an event weight?

Thanks in advance.
Best regards,

Celso

--
You received this question notification because you are an answer
contact for MadGraph5_aMC@NLO.