Errors occured (getting same significance values for all diffrent crosssection rootfiles)

Hello Dr @couet ,
I usually find Errors occured (getting same significance values for all diffrent crosssection rootfiles). This is not occured unless if i have only one value for diffrent
cross scetion. Kindly,please give me a solution for this problem

thanks
Ahmed Sayed

Notes: for more details I attached screenshot from this error , used root files, analysis file.C,analysis file.h and Used Stack


bb.root (28.8 KB)
dataAnalysis.C (10.6 KB)
dataAnalysis.h (83.6 KB)
h_bb.root (16.5 KB)
h90BP1_bb.root (21.0 KB)
h90BP2_bb.root (20.9 KB)
h95BP1_bb.root (19.8 KB)
h95BP3_bb.root (20.4 KB)
h96BP1_bb.root (19.6 KB)
h96BP3_bb.root (20.2 KB)
h96BP2_bb.root (19.7 KB)
h96BP6_bb.root (19.6 KB)
Stack_Hist_Signals.C (11.3 KB)
z_bb.root (20.0 KB)

Hi,

Could you please be more specific about the issue you are encountering?

Best,
D

Of course my Sir.In the beginning i used a BLSSM for eight signals Such that h90B1 ,h90BP2 etc .The rest represent backgrounds that use sm-full model for monte carlo
.The final state for every decay is bb.The channel is proton-proton at 7 TEV at COM . Signals are

p p < h1, h1 < b b that has diffrent parameter
Cards that have same name of root file for simplicity such that h90BP1,h90BP2, etc.And the BGs
are
PP < h , h < b b
PP < z , z < b b
PP < b b / h z

As mentioned below the screenshot significance values are the same for all signals. For more details please kindly ,open
The stack_Signal.C to find followed steps to
Calculate significance.According to observation of signals over backgrounds i calculate signal significance from this relation

Significance =Signal/sqrt(BG) . It is expected that obtain different significance values but unfortunately i get same value for these eight signals . I dont know why ?

Best
A.Sayed

Hi @all experts, 

I usually find Errors occured (getting same significance values for all diffrent crosssection rootfiles). This is not occured unless if i have only one value for diffrent
cross scetion. Kindly,please give me a solution for this problem
.In the beginning i used a BLSSM for eight signals Such that h90B1 ,h90BP2 etc .The rest represent backgrounds that use sm-full model for monte carlo
.The final state for every decay is bb.The channel is proton-proton at 7 TEV at COM . Signals are

p p < h1, h1 < b b that has diffrent parameter
Cards that have same name of root file for simplicity such that h90BP1,h90BP2, etc.And the BGs
are
PP < h , h < b b
PP < z , z < b b
PP < b b / h z

As mentioned below the screenshot significance values are the same for all signals. For more details please kindly ,open
The stack_Signal.C to find followed steps to
Calculate significance.According to observation of signals over backgrounds i calculate signal significance from this relation

Significance =Signal/sqrt(BG) . It is expected that obtain different significance values but unfortunately i get same value for these eight signals . I dont know why ?

Best
A.Sayed

bb.root (28.8 KB)
dataAnalysis.C (10.6 KB)
dataAnalysis.h (83.6 KB)
h_bb.root (16.5 KB)
h90BP1_bb.root (21.0 KB)
h90BP2_bb.root (20.9 KB)
h95BP1_bb.root (19.8 KB)
h95BP3_bb.root (20.4 KB)
h96BP6_bb.root (19.6 KB)
h96BP1_bb.root (19.6 KB)
h96BP2_bb.root (19.7 KB)
Stack_Hist_Signals.C (11.3 KB)
z_bb.root (20.0 KB)

I might be mistaken, but based on your description, it seems that your issue isn’t caused by any specific ROOT tools or classes. Instead, it appears to be related to the algorithm and computational techniques you’re using. If you can identify a specific problem or error directly related to ROOT tools, please send us a small reproducible example highlighting the issue.

Ok, in short . please ,kindly Opeth this file (Signal_stack.C ) the you find this computational values that made me errors.I will send you a screenshot to understand me

thanks
A.Sayed

Stack_Hist_Signals.C (9.42 KB)

Error is found as shown black screenshot (same significance value for diffrent signals). i dont know is this error related to root or a certain mistake in
computational values.Please check this

thanks
Ahmed

The file h96BP3_bb.root is missing when I execute your macro.

im sorry Doctor. I will send you
h96BP3_bb.root (20.2 KB)
now

Your code gives me:

root [0] .x Stack_Hist_Signals.C
Error in <HandleInterpreterException>: Trying to access a pointer that points to an invalid memory address.
Execution of your code was aborted.
In file included from input_line_8:1:
/Users/couet/Downloads/Stack_Hist_Signals.C:87:7: warning: invalid memory pointer passed to a callee:
      hmc[0][i]->SetLineColor(kYellow);
      ^~~~~~~~~
root [1] 

I have modified the macro as follow:

  • the ROOT files paths were not correct in my context
  • I removed useless code
  • I indented the code.

here it is:

Stack_Hist_Signals.C (8.2 KB)

But I get same Significance values


Stack_Hist_Signals.C (11.3 KB)

Your example does not work for me. I get this error:

Processing Stack_Hist_Signals.C...
Error in <HandleInterpreterException>: Trying to access a pointer that points to an invalid memory address.
Execution of your code was aborted.
In file included from input_line_8:1:
/Users/couet/Downloads/Stack_Hist_Signals.C:142:6: warning: invalid memory pointer passed to a callee:
     hmc[0][i]->SetLineColor(kYellow);
     ^~~~~~~~~

Hi,

the issue is that the numerators of all your significances

double S1= Ns1/sqrt(Nb1+Nb2+Nb3);
double S2= Ns2/sqrt(Nb1+Nb2+Nb3);
double S3= Ns3/sqrt(Nb1+Nb2+Nb3);
double S4= Ns4/sqrt(Nb1+Nb2+Nb3);
double S5= Ns5/sqrt(Nb1+Nb2+Nb3);
double S6= Ns6/sqrt(Nb1+Nb2+Nb3);
double S7= Ns7/sqrt(Nb1+Nb2+Nb3);
double S8= Ns8/sqrt(Nb1+Nb2+Nb3);

are the integrals of some histograms

double Ns1=hmc[0][4]->Integral();
double Ns2=hmc[1][4]->Integral();
double Ns3=hmc[2][4]->Integral();
double Ns4=hmc[3][4]->Integral();
double Ns5=hmc[4][4]->Integral();
double Ns6=hmc[5][4]->Integral();
double Ns7=hmc[6][4]->Integral();
double Ns8=hmc[7][4]->Integral();

after you have scaled these histograms in such a way that their integral = 1:

hmc[0][i]->Scale(1/hmc[0][i]->Integral());
hmc[1][i]->Scale(1/hmc[1][i]->Integral());
hmc[2][i]->Scale(1/hmc[2][i]->Integral()); 
hmc[3][i]->Scale(1/hmc[3][i]->Integral());
hmc[4][i]->Scale(1/hmc[4][i]->Integral());
hmc[5][i]->Scale(1/hmc[5][i]->Integral());
hmc[6][i]->Scale(1/hmc[6][i]->Integral());
hmc[7][i]->Scale(1/hmc[7][i]->Integral());
hmc[8][i]->Scale(1/hmc[8][i]->Integral());
hmc[9][i]->Scale(1/hmc[9][i]->Integral());  
hmc[10][i]->Scale(1/hmc[10][i]->Integral());  

No wonder all the values are identical…

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.