MadGraph5_aMC@NLO

reg. cross-section values of a few Wilson coefficients in SMEFTsim

Asked by Rahul Balasubramanian

Dear experts,

         I am currently using the SMEFTsim_A_U35_MwScheme_UFO_v2_1 to generate interference-only samples of single wilson coefficients for the production of Higgs. For instance I was trying to generate ggf Higgs using,

generate g g > h NP = 1 NP^2 == 1 QCD <=2 QED == 1

        Having looked at the vertices.py and coupling.py, the wilson coefficients that enter the ggh vertex are cHG and cHGtil. So subsequently I run the event generation for each of these wilson coefficients using a restrict card, wherein I set everything but the relevant wilson coefficient to 0 and the rest to 0.1. I was expecting non-zero contributions from cHG and cHGtil because they are the only wilson coefficients that enter the vertex.

       For cHG : I get an estimate of : 63.09 +- 0.05481 pb for the cross section
       while for cHGtil : I get a message stating that the survey return zero cross section.

      I would be grateful helping if you could aid in my understanding of this situation and whether the result for cHGtil is expected under the assumptions of the model. I thank you in advance for your time and apologise if I have missed something obvious.

sincerely,
Rahul

Question information

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

Hi,

I do not know what is this model.
However I can provide more details on the issue.
For the moment if you look at the detailled log, you will not have that many detailled explanation.
But in a future version, you will get a faster crash and the detailled log will look like the following:

        1000 points passed the cut but all returned zero
 therefore considering this contribution as zero

In the current version, you should have
 No points passed cuts!
 Loosen cuts or increase max_events 500000

(So it test much more points and does not know find any returning not 0 but does not know if thisi the matrix-element or the cut the reason for that)

Now by putting some print inside the matrix-element evalutaion

           1 ( 0.42974312450064700 , -0.0000000000000000 )
           2 ( 0.0000000000000000 , -77054469694.160233 )
 -> 1 2 ( 0.0000000000000000 , 264909028504.87067 )
 -> 2 1 ( 0.0000000000000000 , -264909028504.87067 )
           1 ( 0.0000000000000000 , 0.0000000000000000 )
           2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 1 2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 2 1 ( 0.0000000000000000 , 0.0000000000000000 )
           1 ( 0.0000000000000000 , 0.0000000000000000 )
           2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 1 2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 2 1 ( 0.0000000000000000 , 0.0000000000000000 )
           1 ( 0.42974312450064700 , -0.0000000000000000 )
           2 ( 0.0000000000000000 , 77054469694.160233 )
 -> 1 2 ( 0.0000000000000000 , -264909028504.87067 )
 -> 2 1 ( 0.0000000000000000 , 264909028504.87067 )
           6
           1 ( 0.42974312450064689 , -0.0000000000000000 )
           2 ( 0.0000000000000000 , -77054469694.160202 )
 -> 1 2 ( 0.0000000000000000 , 264909028504.87048 )
 -> 2 1 ( 0.0000000000000000 , -264909028504.87048 )
           1 ( 0.0000000000000000 , 0.0000000000000000 )
           2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 1 2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 2 1 ( 0.0000000000000000 , 0.0000000000000000 )
           1 ( 0.0000000000000000 , 0.0000000000000000 )
           2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 1 2 ( 0.0000000000000000 , 0.0000000000000000 )
 -> 2 1 ( 0.0000000000000000 , 0.0000000000000000 )
           1 ( 0.42974312450064689 , -0.0000000000000000 )
           2 ( 0.0000000000000000 , 77054469694.160202 )
 -> 1 2 ( 0.0000000000000000 , -264909028504.87048 )
 -> 2 1 ( 0.0000000000000000 , 264909028504.87048 )
           6
           1 ( 0.42974312450064700 , -0.0000000000000000 )
           2 ( 0.0000000000000000 , -77054469694.160233 )
 -> 1 2 ( 0.0000000000000000 , 264909028504.87067 )
 -> 2 1 ( 0.0000000000000000 , -264909028504.87067 )

The line starting by 1 or 2 are the amplitude for the two diagram.
the line starting by -> correspond to contribution added to the interference term.
So indeed the interference is evaluated to zero. You might want to contact the author of the model who knows what this coeficient is if you want a physical reason for such cancelation.

Cheers,

Olivier

On 8 May 2019, at 17:17, Rahul Balasubramanian <<email address hidden><mailto:<email address hidden>>> wrote:

 cHGtil

Revision history for this message
Rahul Balasubramanian (rbalasub) said : 		#2

Thanks for your suggestions and inputs, Olivier.