Note: The track and calorimeter isolation cuts are on the
leptons, as is the opposite sign requirements
An Inclusive Dilepton Analysis
(AIDA)
Sebastian Carron, Mircea Coca,
Mark Kruse
Duke University
(for the CDF-II collaboration.)
Contents
A talk describing our analysis (power point)---->aida_coll_mtng.ppt
(pdf version)---->aida_coll_mtng.pdf
Public
CDF note---->
(you do not need a CDF password)
In Run II a similar,
but somewhat milder excess was observed in the tt cross
section counting expetiment measurement: 9 observed events,
6.8 expected from SM ( a bit less than a 1 sigma effect).
We are thus motivated to make a general study of the dilepton channel,
to
evaluate the degree to which the Standard Model predictions agree with
our
observations.
In this stage of our analysis we develop a methodology to extract simultaneously
the cross sections from the tt, WW and Z->tautau productions. In future
incarnations
of this analysis we intend to extend our scope to make a generalized search
for
exotic processes (Non Standard Model) in the dilepton channel.
Note: The track and calorimeter isolation cuts are
on the
leptons, as is the opposite sign requirements
These cuts essentially just select 2 isolated opposite charged leptons.
Currently we
are able to measure from the dilepton channel the tt, WW and Z->tautau
cross
sections. It must
be noted that we only measure the Z->tautau cross section in the emu
channel, since in the ee and mumu channels we found that a Met Significance
(Met/Root(sum of the Et from the calorimeter towers)) was necessary to
reduce our
otherwise overwhelming Z->ee and Z->mumu Drell Yan background.
There are very few Standard Model processes contributing. We have tt, WW,
Z->tautau as our signal processes, and WZ, ZZ , Wgamma, Drell Yan and fakes
our background processes. Given that all of our Standard Model processes,
apart from 2 leptons only have neutrinos and jets as additional objects,
then
it is natural to consider the Met-Njets distributions as the phase space
where
we perform our fit to the data.
Below we see how, for our signal processes this
Met-Njets phase space nicely
separates our signal contributions. All other contributions in are summed
and
normalized to the expected number of events in the plots below, we also
see the
data distributions for 200 pb-1 of collected data.
emu Met-N jets space
Download --> Met-N jets
in the emu channel
ee Met-N jets space
Download --> Met-N of jets
in the ee channel
mumu Met-N jets space
Download --> Met-Njets
in the mumu channel
Then from these distributions we fit the 3 signal contributions to the
data in the
emu channel and the WW and tt cross sections in the ee+mumu channels (We
can't fit the Z->tautau cross section in the ee and mumu channels). For
our fit we
use likelihood minimization technique described in the next section.
Apart from the increased acceptance over counting experiments, our main
advantage in measuring the cross sections lies in the nice separation attained
in our fitted space.
Below there is a simplified description of the likelihood function we use to do our fit in the
------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------
Download--> Summary of signal and background expectations, and data events
We distinguish between two basic types of systematic errors, those that we include
Bellow we have two tables summarizing the sources and percentual strengths
of the errors
relative to the cross section measurements, for these 2 types of systematic
errors.
Acceptance systematics
Download --> acceptance systematics
table
Also there is a systematic error coming from the
luminosity measurement of about 6%.
Shape systematics
Download --> Main shape systematic contributions
We obtain significant statistical improvements over counting experiments
with this
technique on all measurements
Version 1: Only displaying the cross sections
results with all other contributions fixed
including the number of events found from
the fit.
Download --> cross sections version
1
Version 2: Same as above, but without the fitted number of events
Download --> cross sections version
2
Version 3: Cross sections measured for all processes, both with all
signal processes
floating in the fit simultaneously, and with all but one fixed.
Download --> cross sections version
3
For the emu fit
Download --> emu fit contour for WW and
tt cross sections
Download --> emu fit contour for
the tt and Z->tautau cross sections
Download --> emu fit contour for WW and
Z->tautau cross sections
For the ee+mumu fit
Download --> ee+mumu fit
contour for WW and tt cross sections
Download tables and plots below:
Met-N jets phase space plots (Where we do our fit)
emu phase space:
emu_phasespace.eps
ee phase space ee_phasespace.eps
mumu phase space mumu_phasespace.eps
Monte Carlo expectation and data summary table---> aida_summary_table.eps
3 versions of summarized results with different details
version 1----> aida_results_table_v1.eps
version 2----> aida_results_table_v2.eps
version 3----> aida_results_table_v3.eps
1 sigma confidence level plots (do not include shape systematics)
emu fit 1 sigma confidence plots (for all signals
floating)
contour_emu_ttww.eps
contour_emu_ttztt.eps
contour_emu_wwztt.eps
ee+mumu fit 1 sigma confidence plots (for all signals
floating)
contour_eemumu_ttww.eps
CDF NOTE: 6780 --->Search CDF notes (Need CDF password)
CDF internal web page ----> CDF AIDA web page (Need CDF password)
CDF internal questions and answers web page ---> CDF
AIDA questions and answers (need CDF password)
| Sebastian Carron-
CDF (Duke University)
email: carron@fnal.gov Office (FNAL): (630)840 2724
|