Question about QCD calculations

Hi,

Did you include PDF.
If yes then alpha_s is chosen to be equal to the one used in the PDF fit?

The second typical effect is the running of alpha_s. Which running did you use? (Again we use the one associated to the PDF set)
And to which scale?
I guess that the easiest to compare your MC is
1) check the value of a given PS point (to ensure that you integrate the same function
2) running at fix scale (to avoid any ambiguity with the running)

Cheers,

Olivier

> On 19 Jul 2017, at 15:24, Miguel Rodrigues <email address hidden> wrote:
>
> New question #650028 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/650028
>
> Hi there,
>
> Apologies for the simple question but I am new at this. I am trying to write my own MC event generator for simple processes and I have been comparing my results with Madgraph's. Starting with a simple e+ e- > a > mu- mu+ my result for the cross section agrees with Madgraph and the analytical result. However for a process like u u~ > t t~ QCD=2 QED=0, which should just be replacing alpha_e by alpha_s and multiplying by a color factor 2/9 my results cease to agree with Madgraph's. I have set the values of all relevant parameters to be the ones Madgraph uses, however there is still a very significant difference between the two results. What are the possible reasons for this?
>
> Thanks,
> Miguel
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Hi Olivier,

I am not integrating over PDFs just yet because I wanted to make sure the result for the partonic cross sections were correct first, so I have set the PDF parameters to 0 in Madgraph. I am keeping alpha_s fixed at 1.18e-01, as I assumed Madgraph was too. For example for sqrt(ŝ)=13 TeV, my value for the cross section is 0.0298476 ± 6.66954e-06 pb, whereas the result coming out of Madgraph is 0.01113 +- 1.401e-05 pb, and all it took for madgraph's and my result to disagree were the two seemingly simple changes I mentioned above. How can I check the value for a single PS point in Madgraph if I may ask?

Thanks again,
Miguel

Hi,

Did you set in the run_card, the scale to fix? And the value of the scale to MZ?
(And the value of alpha_s to the value in your code?)

To have a code returning a single phase-space point:
You have all the instructions here:
https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/FAQ-General-4

Cheers,

Olivier

On 19 Jul 2017, at 16:34, Miguel Rodrigues <<email address hidden><mailto:<email address hidden>>> wrote:

Question #650028 on MadGraph5_aMC@NLO changed:
https://answers.launchpad.net/mg5amcnlo/+question/650028

Miguel Rodrigues posted a new comment:
Hi Olivier,

I am not integrating over PDFs just yet because I wanted to make sure
the result for the partonic cross sections were correct first, so I have
set the PDF parameters to 0 in Madgraph. I am keeping alpha_s fixed at
1.18e-01, as I assumed Madgraph was too. For example for sqrt(ŝ)=13
TeV, my value for the cross section is 0.0298476 ± 6.66954e-06 pb,
whereas the result coming out of Madgraph is 0.01113 +- 1.401e-05 pb,
and all it took for madgraph's and my result to disagree were the two
seemingly simple changes I mentioned above. How can I check the value
for a single PS point in Madgraph if I may ask?

Thanks again,
Miguel

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

Hi Olivier,

Thank you, turns out I did not set fixed scale to true, which solved my problem. Thanks!

Best,
Miguel