Re: [ROOT] Binomial y-errors

From: Rene Brun (Rene.Brun@cern.ch)
Date: Thu Mar 06 2003 - 17:42:30 MET


Hi Carl,

An example below;

Rene Brun

void binomial() {
   c1 = new TCanvas("c1","The FillRandom example",200,10,900,600);
   
   gaus1 = new TFormula("gaus1","gaus");
   gaus1->SetParameters(1000,10,3);

   const int nbins = 20;
   h1 = new TH1F("h1","Test random numbers",nbins,0,20);
   h2 = new TH1F("h2","Test random numbers",nbins,0,20);
   h1->Sumw2();
   h2->Sumw2();
   h1->FillRandom("gaus1",8000);
   h2->FillRandom("gaus1",2000);

   TH1F *result = new TH1F("binomial","Test binomial",nbins,0,20);
   result->Divide(h1,h2,1.0,1.0,"b");
   
   result->SetMarkerStyle(21);
   result->Draw();
}



Carl Gwilliam wrote:
> 
> Hi
> 
> I want to plot a histogram with Binomial errors in the y direction, how
> can I do this in root?
> 
> More specifically I am dividing two histograms which plot Ntrig and N
> as the height of each point to get the efficiency as:
> 
> eff = Ntrig/N
> 
> and on this divided histogram I want to plot the errors on each
> point according to the formula:
> 
> eff_error = sqrt(Ntrig/(N*N)*(1-Ntrig/N)).
> 
> At the same time I want to plot the markers as points with x-error bars
> which are the width of the bin.
> 
> Cheers,
> 
> Carl



This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:09 MET