BDT Discriminant Output

For this study I have merged first two preshower bins (pre1=0 && pre2=0 || pre1=0 && pre2>0)

Cuts applied :

(a) z vertex cut -100cm<z<100cm

(b)  Photon and jet are opposite in φ [cos(φΥ – φjet)] < -0.8

(c)  PT of photon > 7GeV

(d) pT of away side jet > 5GeV

(e) Detector eta of the away side jet : ІηjetІ < 0.8

(f) Detector eta of photon 1.08<η<2.0

With these cuts if I look at the pT spectrum of photons, I have 79995 counts in the data and 117888 counts in the MC sample.
So we have a fudge factor of 1.47.

These plots are integrated over all pT bins and first two preshower bins.

I have scaled the prompt photon and qcd background sample to 3.146 pb-1 (data luminosity).

Following 12 variables were used to model the response:

ptBalance, R3x3jet7, R2x2jet7, R2x1jet7, nU, nV, fiveStripU/twentyfiveStripU, fiveStripV/twentyfiveStripV, reslayer1, reslayer2, yieldlayer1, yieldlayer2
 

BDT setup parameters :
 

NTrees=120, MaxDepth=1, AdaBoostBeta=0.72, SeparationType = GiniIndex, nCuts=25, PruneMethod=NoPruning