[Rivet] Leading neutrinos?

[Jonathan Butterworth]

Hi Frank, (and David, see important correction below regarding ATLAS 
W+jets),

On 01/03/2011 12:02, Frank Siegert wrote:

>
> But since I have the convenor of that measurement at hand, I might as
> well ask: What's the reason for not using all neutrinos in your particle
> level Etmiss definition? Is it convenience for parton level calculators
> (in which case I'd say publish both types) or is there any reason why a
> measurement that uses Etmiss from the final state in a way as close to
> the measurement as possible wouldn't be good?

I don't think there is a very strong reason either way, and in fact now 
I remember, if you look closely at our W+jets paper you'll see that both 
neutrinos and muons are excluded from the jets! In this case, the 
behaviour I gave to David is not correct - ALL neutrinos *and muons* 
should be excluded from the interacting final state! Sorry. This is of 
course based on what leaves energy in the calorimeter, so is very close 
to what the detector sees (if you ignore inner tracking and muon 
chambers!).

In ATLAS jet measurements in general (including future W and Z+jets 
measurements) we try and correct back to the hadronic final state based 
on a lifetime cut (typically 10ps).

One has to choose a point at which to make data/theory comparisons and 
the idea of doing it "post hadronisation" was the main one (we didn't 
just make it up, actually, it was discussed at some length in Les 
Houches 2007).

It is not an iron rule (e.g. 10ps cut means B hadrons are treated as 
unstable, which may not be the best thing to do. Of course if they were 
treated as stable, the neutrino correction would be forced for those 
decays. As we don't treat them as stable, if we didn't correct for 
neutrinos, our final state energy would change a lot with the lifetime 
cut...).

There is some login in it, since there are multiplicity and particle 
correlations (e.g. presence of a muon or electron in a jet; B-decays are 
quite well understood for example) which clearly correlate with the 
presence of neutrinos. By correcting using such information we 
potentially reduce the dependence of the measured final state on 
fluctuations in the hadronic final state, so (at the price of 
introducing some small model dependence into the measure itself) we can 
improve the resolution on missing ET from other sources. But in general 
neither the advantage nor the disadvantage is large, as far as I have 
been able to tell.

Of course if there is no killer "right or wrong" answer, the most 
important thing is to state explicitly what is done, which I hope and 
believe at least all ATLAS SM paper do and will continue to do!
(Even if they don't always do exactly the same thing).

Cheers,
Jon

>
> Cheers,
> Frank

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Prof. Jonathan Butterworth,              http://www.hep.ucl.ac.uk/~jmb/
Physics and Astronomy Department                  Tel: +44 20 7679 3444
ATLAS, CERN                                       Tel: +41 22 76  72340
University College London                 Gower St, London WC1E 6BT, UK
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~