[Rivet] Rivet: Accessing intermediate particles

[David Hall]

Hi Andy,

Thanks for such a clear explanation.

Unfortunately, when I check the status of each parent of the electron 
in an event, I find that they are _all_ status 2. This suggests that 
all the electrons are from decays, but I know this shouldn't be the 
case. Two should be from the hard scatter. Is it possible this is 
specific to Sherpa?

Thanks,
David

On Sat 30 Jun 2012 16:21:49 CEST, Andy Buckley wrote:
> Hi David,
>
> It's best not to rely on any particular definition of the "hard process"
> as implemented in different generators: it's rather an arbitrary thing,
> and in fact the distinction would not exist at all if it weren't for the
> limitations of current calculations. The further down the line of
> ME-enhanced parton showers we go, the less the concept makes sense in an
> analysis... and Sherpa is probably the generator furthest down that line
> of development!
>
> Additionally, matching on momenta between the internal and final state
> particles in an event record is *very* fragile: best not to go there.
>
> In fact the parton showering will not emit more *leptons*, but hadronic
> decays will. Fortunately the HepMC status codes can be used to
> distinguish particles from decays from those produced in the rest of the
> event: if you use an IdentifiedFinalState (or similar) in Rivet to find
> all the final state leptons that you are interested in, then loop over
> them and ass each one's HepMC GenParticle (via the genParticle() method)
> what the status code of its parent was: if the parent status is 2, then
> it's a lepton from a hadronic decay; if it's something else (0, 3, or >
> 10) then it's the sort of "prompt" lepton that you're interested in.
>
> Note, however, that distinguishing leptons from the gamma* from those
> produced in leptonic decay of the W is not physically meaningful if done
> using truth record information about parentage... but a
> reconstruction-type approach, using windowing mass(l+nu) around the W
> mass would be robust and physical since it's based on real observable
> quantities (the neutrino momentum may not be practically observable with
> real detectors, but there's no reason not to use it in Rivet!)
>
> Cheers, and I hope that helps,
> Andy
>
>
> On 29/06/12 11:02, David Hall wrote:
>> Hello,
>>
>> As a bit of background, I am using Sherpa to output a HepMC event file
>> and then Rivet to perform an analysis on this file. I am studying the
>> W+gamma* process, and need to identify the leptons that are coming from
>> the gamma*. When parton showering is performed, additional leptons are
>> produced and it becomes difficult to be certain which leptons are from
>> the hard process.
>>
>> I have heard that after the matrix element calculation, Sherpa outputs
>> the four-momenta from the hard scatter as additional information in the
>> HepMC file (perhaps as another GenVertex?). So I was thinking, if I had
>> access to this information, I could try to match the 4-momenta of the
>> final state particles to those from the hard process in order to
>> identify the leptons I need.
>>
>> I was wondering if there is anything in Rivet to provide access to this
>> information? I can only find the FinalState particles currently. I
>> understand the philosophy behind Rivet not providing access to the
>> intermediate particles, but I believe that looking for the particles
>> from the hard process in this way is not _so_ bad. :-)
>>
>> Thanks,
>> David
>>
>> _______________________________________________
>> Rivet mailing list
>> Rivet at projects.hepforge.org
>> http://www.hepforge.org/lists/listinfo/rivet
>>
>
>