Cross section is strongly dependent on xqcut when jet matching with manual decay

[Jake Fennick]

When performing jet matching, the cross section is supposed to be relatively independent of the parameter xqcut. This seems to be true for processes when no decay is specified. However, when an explicit decay chain is specified, the cross section is strongly dependent on xqcut, and decreases monotonically.

See attached process files for the exact processes. (All cross sections are in femtobarns.)

p p > t t~ + jets, no decay
xqcut cross section
20 115801
40 111458
60 112994
80 113959
100 114071

p p > t t~ + jets, with decay
xqcut cross section
20 23702.5
40 3581.28
60 714.48
80 74.6443
100 0.0012087

p p > w + jets, no decay
xqcut cross section
20 7.67588e+07
40 7.51619e+07
60 7.38917e+07
80 7.3294e+07
100 7.30013e+07

p p > w + jets, with decay
xqcut cross section
20 3.92864e+07
40 4.39186e+06
60 42268
80 6976.7
100 38.1674

[Olivier Mattelaer (olivier-mattelaer)]

Hi,

Do you have auto_ptj_mjj on True?
Do you have cut_decay on True?
Do you have your W decaying in jet?

If you have three yes, then this is normal, since the auto_ptj_mjj change the ptj to the value of xqcut.
In that case you also have a cut on the jet coming from the W and this make a dependence in your cross-section.

Cheers,

Olivier

On 10 Apr 2015, at 03:37, Jake Fennick <email address hidden> wrote:

> Public bug reported:
>
> When performing jet matching, the cross section is supposed to be
> relatively independent of the parameter xqcut. This seems to be true
> for processes when no decay is specified. However, when an explicit
> decay chain is specified, the cross section is strongly dependent on
> xqcut, and decreases monotonically.
>
> See attached process files for the exact processes. (All cross sections
> are in femtobarns.)
>
> p p > t t~ + jets, no decay
> xqcut cross section
> 20 115801
> 40 111458
> 60 112994
> 80 113959
> 100 114071
>
> p p > t t~ + jets, with decay
> xqcut cross section
> 20 23702.5
> 40 3581.28
> 60 714.48
> 80 74.6443
> 100 0.0012087
>
> p p > w + jets, no decay
> xqcut cross section
> 20 7.67588e+07
> 40 7.51619e+07
> 60 7.38917e+07
> 80 7.3294e+07
> 100 7.30013e+07
>
> p p > w + jets, with decay
> xqcut cross section
> 20 3.92864e+07
> 40 4.39186e+06
> 60 42268
> 80 6976.7
> 100 38.1674
>
> ** Affects: mg5amcnlo
> Importance: Undecided
> Status: New
>
> ** Attachment added: "mg5_processes.zip"
> https://bugs.launchpad.net/bugs/1442442/+attachment/4371100/+files/mg5_processes.zip
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1442442
>
> Title:
> Cross section is strongly dependent on xqcut when jet matching with
> manual decay
>
> Status in MadGraph5_aMC@NLO Generator:
> New
>
> Bug description:
> When performing jet matching, the cross section is supposed to be
> relatively independent of the parameter xqcut. This seems to be true
> for processes when no decay is specified. However, when an explicit
> decay chain is specified, the cross section is strongly dependent on
> xqcut, and decreases monotonically.
>
> See attached process files for the exact processes. (All cross
> sections are in femtobarns.)
>
> p p > t t~ + jets, no decay
> xqcut cross section
> 20 115801
> 40 111458
> 60 112994
> 80 113959
> 100 114071
>
> p p > t t~ + jets, with decay
> xqcut cross section
> 20 23702.5
> 40 3581.28
> 60 714.48
> 80 74.6443
> 100 0.0012087
>
> p p > w + jets, no decay
> xqcut cross section
> 20 7.67588e+07
> 40 7.51619e+07
> 60 7.38917e+07
> 80 7.3294e+07
> 100 7.30013e+07
>
> p p > w + jets, with decay
> xqcut cross section
> 20 3.92864e+07
> 40 4.39186e+06
> 60 42268
> 80 6976.7
> 100 38.1674
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1442442/+subscriptions