MadGraph5_aMC@NLO

Issue with cross section computation for a SUSY process

Asked by HH

Dear MadGraph experts,

For a SUSY (MSSM) analysis, I need to simulate signal events. I am using MadGraph5 with the MSSM model written by Benjamin Fuks (default MSSM model in MadGraph) and a parameter card generated by FeynHiggs-2.19.0. During the generation of events, I encountered a problem with the cross sections computed by MadGraph5. I was hoping you could help me find a solution.

The process under study is p p > h01 z, h01 > b b~, where h01 is the light CP-even Higgs boson (the model has 5 Higgs bosons, 2 CP-even neutral Higgs bosons h01,h02, 1 pseudoscalar Higgs boson A0 and 2 charged Higgs bosons H+/H-). In this process, the produced Higgs boson decays into a pair of bottom quarks. The Higgs masses are assumed to be: M_h01 = 66 GeV, M_h02 = 123 GeV, which means that the heavy CP-even Higgs boson is assumed to be the SM-like Higgs boson in the model. The problem I encountered is that the cross section of this process is not consistent with the cross section of the process p p > h01 z. It is expected that: Xsec(p p > h01 z, h01 > b b~) = Xsec(p p > h01 z)*BR(h01 > b b~) with BR(h01 > b b~)~0.8. But the Xsections computed by MadGraph5 are: Xsec(p p > h01 z)~0.43 pb, Xsec(p p > h01 z, h01 > b b~)~0.82, which is clearly incorrect.

I would appreciate it if you could help me find a solution for this problem.

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

Did you check the FAQ?
FAQ #2442: “why production and decay cross-section didn't agree.”.

Revision history for this message
HH (hh-madgrapher) said : 		#2

Thanks for your answer. Actually, I had read that FAQ.

As an update, I found out that the main problem is the difference between the decay widths computed by FeynHiggs and the MSSM model for h01 Higgs boson. Both the total and partial widths for the h01 Higgs decay are different from those calculated by FeynHiggs. In fact, the widths calculated by MadGraph are larger than those computed by FeynHiggs at 2-loop level by a factor of 2. I always expect the width calculated by MG to be smaller since it is calculated at the tree-level.

Here are the flag settings I used in FeynHiggs:
mssmpart = 4 # full MSSM [recommended]
higgsmix = 2 # 2x2 (h0-HH) mixing in the neutral Higgs sector
p2approx = 4 # no approx., UHiggs evaluated at p^2 = 0 [recommended]
looplevel = 0 (and 2) # include 2L corrections
loglevel = 3 # NNLL resummation (for large MCha,MNeu,MGlu,MSUSY)
runningMT = 1 # use running top mass (SM MSbar 2L)
botResum = 1 # resummed 1L-MB in 1L/2L Higgs self-energy corrections

The settings in the MSSM model are the default settings.
Since the values of the input parameters in FeynHiggs and the MSSM model are the same, I suspected that the theoretical frameworks have different assumptions. I would appreciate it if you could let me know what you think.

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

Hi,

I personally do not expect a two-loop computation to be larger than a tree level computation.
For cross-section for example, K factor are typically larger than one, but this assume that you pick the same value of alphas for both combination, while in practise LO computation takes a larger value of alphas than NLO computation, so they are actually a compensating effect. But even if you neglect that compensating effect, they are no reason for the K-factor to be always larger than one.

For width computation, I have close to zero experience on NLO and NNLO computation, so cannot really comment on those.
Now a factor of 2 with FeynHiggs sounds a quite large difference, I found that intriguing, maybe running FeynHiggs at tree level to check the compatibility with MadGraph and check partial width by partial width to understand why the difference is so large.

Cheers,

Olivier

Can you help with this problem?

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

To post a message you must log in.