Replicating Pass 6 Cuts
I've gone back to Eric's old page with the hope of picking up where he left off. Of course, this leads to the always enjoyable task of replicating the work that he has already done. The working directory for the previous analysis is:
/afs/slac/u/ek/echarles/vol2/glast_workdirs/GR_v15r50p3
From there, I have been able to more or less reconstruct the analysis that Eric was performing. Building my own version of GR v15r50p3, I used the runme.csh and runme_ag.csh scripts (with minor modifications) in order to generate a set of background and allGamma MC. Key points about these MC simulations:
- The bkg sims are on orbit using a mix of particle types...
- Other ???
Looking in gamma_ana.txt, I found the exact cuts that Eric was using. Namely, the prefilters that were being applied before the cut efficiencies were evaluated. I've listed them here for completeness.
nSig |
nBkg |
|---|---|
~9x10^5 |
~4x10^6 |
Name |
Purpose |
Gamma Efficiency |
Bkg Efficiency |
Cut |
|---|---|---|---|---|
IrrFilter |
Remove irreducible bkg events |
0% |
13% |
((McZDir < -.2 && McId < 20) | (McZDir < .1 & McId > 20)) && McAcdZEnter > 100 && McCharge !=0 && McTHPosHitOthers < 6 |
TkrCalCore |
Take only "good" events |
6.2% |
2.2% |
TkrNumTracks>0 && CalEnergyRaw > 5 && CalCsIRLn>4 && CTBCORE>0.1 |
AcdEngFilter |
Only events with moderate signal in the ACD |
0% |
47% |
(AcdTotalEnergy+AcdRibbonEnergy<0.05) && McZDir > 0 |
Prefilter |
Only select relevant events |
6.2% |
1.2% |
(TkrCalCore && !(IrrFilter) && !(AcdEngFilter) |
nSig |
nBkg |
|---|---|
55015 |
48614 |
Name |
Purpose |
Gamma Efficiency |
Bkg Efficiency |
Cut |
|---|---|---|---|---|
BasicTileCut |
|
95.2% |
7.85% |
(Tkr1SSDVeto == 0 && AcdTkr1ActiveDist > -16 && AcdTkr1ActDistTileEnergy > .4) |
RibbonCut |
|
94.9% |
7.43% |
(AcdRibbonActDist > -(2 +350/sqrt(max(20,CTBBestEnergy))) && Tkr1SSDVeto < 3 && AcdRibbonEnergy > .05 ) |
TotalTileEnergyCut |
|
89.3% |
1.17% |
(AcdTotalTileEventEnergyRatio > .8 || (AcdTkr1ActiveDistENorm > -300 && AcdTotalTileEventEnergyRatio > max(.005, .1 - .0001*AcdTkr1ActiveDistENorm))) |
CornerCut |
|
88.4% |
1.15% |
((Tkr1LATEdge/1.5) ^ 2 + (AcdCornerDocaENorm - 10)^ 2 < 3800 || (Tkr1LATEdge < 80 && abs(AcdCornerDocaENorm-2) < 4)) && Tkr1SSDVeto < 3 |
TileEdgeCut |
|
88.3% |
1.15% |
(abs(AcdTkr1ActiveDistENorm) < 15 && AcdTotalTileEventEnergyRatio > .005) |
These numbers seem to agree well with Eric's results. However, it naively seems like there are still some photons sneaking into the bkg sample. This is because selecting !(IrrFilter) will return True if McCharge == 0 (leaving about ~50 gamma events out of ~550). Putting a cut on McCharge, we now get a background rejection of 1.03%.