Add a cut at the lumi block level : if Nevt of lumi block > 1000 and SHIFTmetb of the lumi block > 10 => exclude the run.
This removes 9 runs more.
Keep 85 % of runs (89.2 % of lumi blk and 89.0 % of events)
Presentations at CALGO meetings
6 oct 03 :
MET status
21 oct 03 : MissingET
corrections
18 nov 03 : Status
of MET certification
3 dec 03 : MET
Certification
3 dec 03 : MET
and d0correct
MET Certification on p14
334
runs
studied between 178309 and 180956
47490 lumi blocks => 106.9
Mevts (52.6 pb-1)
• In the « official»
Physics Runs database ? 337 runs listed
(http://www-d0.fnal.gov/runcoor/d0_private/operations/stores_runs.html)
• 326 in common
• 11 are in
the db but are zeroB or not in tmb
• 24
runs are excluded because
flagged bad by Slava or after reading the logbook
(43698 lumi blocks / 98.0 Mevts
/ 48.5 pb-1)
• Studies done at cc-in2p3 on thumbnails
Run selection
For each run we run on all events
and get the mean values and RMS of METBx, METBy,SETB and MET-T42x, MET-T42y,SET-T42
distributions.
Definitions :
METB = MET computed with cells
not belonging to the CH (Ecell>0, layer<15) and with the CH fraction
of JCCB jets.
MET-T42 = METC (Ecell>0) – T42
cells
Variables computed with mean values
of distributions of all events in a run.
Cuts
SHIFT-metbxy < 4 GeV
RMS-metbxy < 22 GeV
Mean-setb > 80 GeV
Nevts > 1000
We keep 276
runs => 88.7 % (92.7 % of lumi blk and 92.9 % of
events)
But still large fluctuations of <METBx> and
<METBy> per lumi blocks
<METBx> (left) and <METBy> (right) for each luminosity block as function of the number of events in the lumi block after the run selection :
Add a cut at the lumi block level :
if Nevt of lumi block > 1000 and SHIFTmetb of the lumi block > 10
=> exclude the run.
This removes 9 runs more.
Keep
85 % of runs (89.2 % of
lumi blk and 89.0 % of events)
Luminosity block selection
For each lumi block we run on all
events and get the mean values and RMS of METBx, METBy,SETB and MET-T42x,
MET-T42y,SET-T42 distributions.
Cuts at the lumi block level :
SHIFT-metbxy < 4 GeV
RMS-metbxy < 22 GeV
Mean-setb > 80 GeV
Nevts > 500
With this selection we
keep 40597 lumi blocks / 93.2 Mevts
and we save 92.9
% of lumi bks or 95.1 % of evts
METBx
for each lumi block
METBy for each lumi block
Before selection
Before selection 
After selection
After selection
<METB> for each lumi block after selection RMS METB for each lumi block after selection
missingET Package documentation
Description of the corrections
Currently in the new missingET
package (v00-06-05) MET is corrected for jet
energy scale and muons.
The em-scale is being implemented
now and not yet available.
In the future it will be also corrected
for unclustured energy.
Old MET variables will still be
in the missingET chunk for backward compatibility and new variables of
corrected MET are added.
The package missingET is called
by d0correct afterthe
other Id packages and MET corrections come from these packages.
The missing transverse energy corrected
is META or METB (generically called MET below).
It's determined on cells with energy>0
and not belonging to the CH. The CH fraction of good JCCA and JCCB jets
is also taken (META and METB refer to JCCA/B).
CORRECTIONS
*
For the jet energy scale, there is a list of good certified jets provided
by the method isGoodJet(*jet).
For each of these jet,
there is a correction to apply to MET given by :
jetcorr:forMET.
The fact that the jet is
matched with a muon is taken into account to determine the JES correction
applied to MET.
=>
Note that even if the jet is matched with a muon, this correction will
only be for the jet part and at this step, MET is not corrected for the
muon.
* For
the muons, there is a list of good certified medium
muons provided by the method isGoodMuon(*muon).
The correction is made simply by
subtracting the transverse momentum of each muon to MET.
Note that the correction is applied
for all medium muons (track matched or not).
* For
the electro-magnetic scale, there is a list of good certified em
objects provided by isGoodEM(*em).
There is a matching made between
each em object and the jets. For the em objects which are not jets, an
em-scale correction is applied.
The corrected MET is returned by the following methods :
getMETACorrCALO
-> MET
corrected for JES of JCCA jets
and for electrons
getMETACorrCALOxy
-> METx , METy
(and for em-scale in next version)
getSETACorrCALO
-> SET
getMETBCorrCALO
-> MET
corrected for JES of JCCB jets
and for electrons
getMETBCorrCALOxy
-> METx , METy
(and
for em-scale in next version)
getSETBCorrCALO
-> SET
getMETACorrCALOMU
-> MET
corrected for JES of JCCA jets,
for electrons and muons
getMETACorrCALOMUxy
-> METx , METy (and
for em-scale in next version)
getSETACorrCALOMU
-> SET
getMETBCorrCALOMU
-> MET
corrected for JES of JCCB jets,
for electrons and muons
getMETBCorrCALOMUxy
-> METx , METy (and
for em-scale in next version)
getSETBCorrCALOMU
-> SET
Sample WZ, Z->ee , 117pb-1 , 12000 evts, p14TMBfixer01
+ d0correct
MET corrections for JCCA (blue)
and JCCB (red) for respectively CH, JES, bad jets,
em-scale and muons
Requiring no jets
METC (black), METD (green), METE
(purple) with METE = METC -T42 corrected
META (blue) and METB (red)
Uncorrected Corrected
