BDT numbers sorted into preshower bins (New)

 BDT numbers sorted into preshower bins

(I) Overtraining Check Plots

(a) First preshower bin                                                                               (b) Second preshower bin     

 

 

(c) Third preshower bin                                                                             (d) Fourth preshower bin

  

 

II) Background rejection vs signal efficiency plots

(a) First preshower bin                                                                                (b) Second preshower bin

   

 

(c) Third preshower bin                                                                            (d) Fourth preshower bin

  

 

(III) Efficiency plots

(a) First preshower bin                                                                               (b) Second preshower bin

  

 

(c) Third preshower bin                                                                             (d) Fourth preshower bin

  

 

Table I

Preshower bin #signal, #background Optimal-cut S/sqrt(S+B) Nsig Nbkg EffSig Effbkg
First (pre1=0,pre2=0) 206,116 -0.4031 12.02 206 88 0.99 0.75

Second

(Pre1=0, Pre2>0)

416, 586 -0.0763 15.6 365 183 0.88 0.31
Third (0<Pre1<4MeV) 1020, 3198 -0.1099 21.97 860 673 0.84 0.21

Fourth

(4<Pre1<10MeV)

874, 5610 -0.0672 17.88 542 379 0.62 0.068

Table 2 : BDT Setup

Preshower bin NTrees nEventsMin MaxDepth AdaBoostBeta nCuts
First 50 10 1 0.25 2
Second 150 10 1 0.05 2
Third 120 10 1 0.48 5
Fourth 50 20 1 0.76 20
 

 

NOTE: I have retained preshower and postshower info as part of the fit in these distributions wherever possible. Please see the attached files for input variables.