[Rivet] Validation & tuning phone meeting, Monday 29th Sept, 4pm (UK)

[Lars Sonnenschein]

Just a short heads up:

I have put two new plots at:

http://www-d0.fnal.gov/~sonne/D0_2001_S4674421/

which I made to show, that using selection cuts and isolation criteria on 
reconstructed objects don't make much of a difference in comparison to 
the partonic boson analysis which is in the Rivet head version.

Having said that, I will commit into the head version the
partonic boson analysis as used for the plots above (temporarily).
Because the reconstructed analysis version contains much more code
and I want to find the needle before growing the stack of hay.

Concerning the plots: The Pythia simulation is scaled to the data.
The ratio is obtained by the scaled histo's, using ROOT's 
histogram division function. Without scaling to data the ratio
is by about 20% lower.

I checked with the very first bin value in the histo's to be divided that 
ROOT is doing the right thing. The ratio histo from Rivet is about a 
factor of two lower than the Rivet answer. Also the errors come out much 
smaller (too small I suspect). So I have to check the 
dividing function in Rivet. That is one thing.
(all the plots I put on the web contain the ratio determined 
retroactively by ROOT)
 
The other thing is that when I run HerwigJimmy from the cern afs 
cluster it's trying events but I never see the Rivet 100 events control 
message, then after about 20 minutes my interactive job gets killed from 
a deamon. So there are no accepted events.

I don't know what's wrong and so I have to repeat this HerwigJimmy 
exercise on my laptop.

Meanwhile I will run just Herwig (without Jimmy) via cern afs and try 
Emily's parameter settings. My Herwig parameter file looks than for Z boson production like:

#ATLAS Tune from Emily
PTRMS = 1.6
VQCUT = 0.45
QCDLAM = 0.14

MAXER = 1000
IPROC = 1351 
EMMIN = 60
#--beam1 PROTON --beam2 ANTIPROTON --mom1 300 --mom2 300
IPRINT = 1
MAXPR = 1

So far

	Lars



Having said that, I will commit into the head version the

On Thu, 9 Oct 2008, Emily Nurse wrote:

> Hi Lars,
> 
> It doesn't look like the HERWIG plots are normalised to the data, is this
> right?
> Could you do that? Otherwise its hard to see how well the shapes agree.
> 
> I'm surprised by how badly the MC is describing the low pT region.
> For the HERWIG parameters you could try:
> 
> PTRMS = 1.6
> VQCUT = 0.45
> QCDLAM = 0.14
> 
> but these were tuned without Jimmy so may not be the best fit when you run
> with Jimmy.
> 
> I'm surprised that the ratio plot is still going in the opposite direction to
> the data even after scaling the Z pT by Mw/Mz.
> The Z pT distribution peaks higher than the W pT because of the higher Z mass
> (more phase space available for gluon radiation at that energy). So without
> the scaling I would expect the W : Z ratio to decrease with pT. With the
> scaling it should go slightly the other way (I think because alpha_s is weaker
> at Q=Mz than at Q=Mw). This is what I saw when I did this study with HERWIG
> and what the data sees. So something funny is going on.
> 
> Could I take a quick look at your running parameters/switches and the routine?
> > 
> > Concerning the overal normalisation there is an ambiguity:
> > boson->tau->electron.
> > I didn't switch this on, since the papers only talk about boson->e('s)
> > without intermediate tau. This would increase the xsec roughly by 1/3, but
> > the pT spectrum of the electrons from taus is also a little bit different.
> > So if I would switch boson->tau on, it depends on how the experiment has
> > corrected (to boson->e('s) or to boson->e('s) + boson->tau('s)->e('s)).
> > 
> 
> You should run only Z->ee and W->enu, no tau final states. The number of
> Z->tautau and W->taunu events passing the cuts will be small (the electron pT
> spectrum is much softer) and the experiments will have (should have) included
> it as a background.
> 
> Cheers,
> Emily.
> 
> > 
> > 
> > On Mon, 6 Oct 2008, Emily Nurse wrote:
> > 
> > > Hi Lars,
> > > 
> > > I think you will need to correct for the branching ratio difference as
> > > this is
> > > a ratio of production cross sections , so your formula
> > > > R_th = dsigma/dpt(W) / dsigma/[d(pt(Z)*Mw/Mz)] *
> > > > B(Z->ee)/B(W->enu)
> > > 
> > > looks correct - although its been a long time since I read the paper. If
> > > the
> > > normalisation looks off lets re think it.
> > > 
> > > Cheers,
> > > Emily.
> > > 
> > > On 6 Oct 2008, at 09:48, Lars Sonnenschein wrote:
> > > 
> > > > Hello Emily
> > > > thank you for your feedback.
> > > > Indeed the paper has two formulas for the ratio:
> > > > a theoretical one which corresponds to your formula below and an
> > > > experimental one which looks like:
> > > > 
> > > > R_exp(pT) = [dsigma/dpt(W) / dsigma/dpt(Z)] * Mw/Mz * B(Z->ee)/B(W->enu)
> > > > 
> > > > where Mw/Mz = 0.88 and B(Z->ee)/B(W->enu) = 0.3 roughly.
> > > > The later formula I use for the ratio, i.e. scale the diff. xsec. ratio
> > > > by the constant factors given above.
> > > > 
> > > > Now I could scale the Z pT before filling the histo according to the
> > > > theoretical formula, but that would not alter the W pT spectrum.
> > > > 
> > > > Though I will try to see the difference in the ratio.
> > > > 
> > > > Now, since I select electrons (and  a neutrino in case of W) should the
> > > > formula look like R_th = dsigma/dpt(W) / dsigma/[d(pt(Z)*Mw/Mz)] *
> > > > B(Z->ee)/B(W->enu)
> > > > to take the branching ratio differences into account?
> > > > 
> > > >      Lars
> > > > 
> > > > 
> > > > On Mon, 6 Oct 2008, Emily Nurse wrote:
> > > > 
> > > > > Hi Lars,
> > > > > 
> > > > > Its hard to see how off the W pT plot is, can you compare it to the
> > > > > data
> > > > > so we
> > > > > can see?
> > > > > As for the ratio, its more obvious that its going the wrong way
> > > > > compared
> > > > > to
> > > > > the data.
> > > > > In the paper they are measuring dsigma/dpt(W) / dsigma/[dpt(Z)*Mw/Mz]
> > > > > ie/ they scale the Z pT by the ratio of the W and Z masses, see page
> > > > > 20
> > > > > and 21
> > > > > of my 1st year PhD transfer report for more details, and comparisons
> > > > > of
> > > > > the
> > > > > data with Herwig:
> > > > > http://www.hep.man.ac.uk/u/emily/1styear.ps
> > > > > 
> > > > > It should also be explained in the paper. Are you doing this in the
> > > > > routine?
> > > > > 
> > > > > Cheers,
> > > > > Emily.
> > > > > 
> > > > > 
> > > > > On 6 Oct 2008, at 08:19, Lars Sonnenschein wrote:
> > > > > 
> > > > > > Hi
> > > > > > I have put some plots for the D0 2001 WpT <-> ZpT analysis at
> > > > > > http://www-d0.fnal.gov/~sonne/D0_2001_S4674421/
> > > > > > 
> > > > > >      Lars
> > > > > > 
> > > > > > 
> > > > > > On Fri, 3 Oct 2008, Andy Buckley wrote:
> > > > > > 
> > > > > > > Hi everyone,
> > > > > > > 
> > > > > > > Our next Rivet etc. phone meeting will take place on Monday at 2pm
> > > > > > > (UK
> > > > > > > time). Note that it's earlier than last week - this will be the
> > > > > > > normal
> > > > > > > time from now on.
> > > > > > > 
> > > > > > > Once again, we'll be using the DESY phone conference system: phone
> > > > > > > 
> > > > > > > +49 40 8998 1390
> > > > > > > 
> > > > > > > then enter conference code 74838 ("rivet") followed by #.
> > > > > > > 
> > > > > > > From the UK, Telediscount can be useful if your office restricts
> > > > > > > international dialling: 0844 861 3535.
> > > > > > > 
> > > > > > > Have a good weekend!
> > > > > > > Andy
> > > > > > > 
> > > > > > > _______________________________________________
> > > > > > > Rivet mailing list
> > > > > > > Rivet at projects.hepforge.org
> > > > > > > http://www.hepforge.org/lists/listinfo/rivet
> > > > > > > 
> > > > > > 
> > > > > > -- 
> > > > > > 
> > > > > > 
> > > > > > ________________________________
> > > > > > Lars Sonnenschein
> > > > > > ________________________________
> > > > > > Home Institution:
> > > > > > PH/TH 53/1-052
> > > > > > CERN
> > > > > > CH-1211 Geneve 23
> > > > > > Switzerland
> > > > > > Tel.:+41(22)767-2801
> > > > > > --------------------------------
> > > > > > ________________________________
> > > > > > FNAL:
> > > > > > D0, PK151
> > > > > > Mailstop #352
> > > > > > Fermilab, P.O.Box 500
> > > > > > Batavia, IL 60510-500
> > > > > > USA
> > > > > > Tel.: +1(630)840-8740
> > > > > > ________________________________
> > > > > > _______________________________________________
> > > > > > Rivet mailing list
> > > > > > Rivet at projects.hepforge.org
> > > > > > http://www.hepforge.org/lists/listinfo/rivet
> > > > > 
> > > > > 
> > > > 
> > > > -- 
> > > > 
> > > > 
> > > > ________________________________
> > > > Lars Sonnenschein
> > > > ________________________________
> > > > Home Institution:
> > > > PH/TH 53/1-052
> > > > CERN
> > > > CH-1211 Geneve 23
> > > > Switzerland
> > > > Tel.:+41(22)767-2801
> > > > --------------------------------
> > > > ________________________________
> > > > FNAL:
> > > > D0, PK151
> > > > Mailstop #352
> > > > Fermilab, P.O.Box 500
> > > > Batavia, IL 60510-500
> > > > USA
> > > > Tel.: +1(630)840-8740
> > > > ________________________________
> > > 
> > > 
> > 
> > -- 
> > 
> > 
> > ________________________________
> > Lars Sonnenschein
> > ________________________________
> > Home Institution:
> > PH/TH 53/1-052
> > CERN
> > CH-1211 Geneve 23
> > Switzerland
> > Tel.:+41(22)767-2801
> > --------------------------------
> > ________________________________
> > FNAL:
> > D0, PK151
> > Mailstop #352
> > Fermilab, P.O.Box 500
> > Batavia, IL 60510-500
> > USA
> > Tel.: +1(630)840-8740
> > ________________________________
> 
> 

-- 

 
________________________________
Lars Sonnenschein    
________________________________
Home Institution:
PH/TH 53/1-052 
CERN
CH-1211 Geneve 23
Switzerland
Tel.:+41(22)767-2801
--------------------------------
________________________________
FNAL:
D0, PK151
Mailstop #352
Fermilab, P.O.Box 500
Batavia, IL 60510-500
USA
Tel.: +1(630)840-8740
________________________________