Hi Hajime,
Thanks for reporting this problem when computing the errors
in TH1::Multiply(TF1*). Now fixed in CVS
Rene Brun
Hajime Nanjyo wrote:
>
> Dear Rooters,
>
> I performed histogram multiplication by 0.1,
> using the member function "Multiply(TF1* h1, Double_t c1 = 1)".
>
> The same operation could be done in two ways as follows.
>
> constant function * constant,
> 1) 1.0 * 0.1
> 2) 0.1 * 1.0
>
> The bin contents of histos after both methods are same,
> but the errors are not same.
> Why?
>
> ROOT Version 3.03/08 was used.
> A simple script is listed below.
> {
> //////////////////////////////////////////////////
> //<Prepare two same histogram with errors>
> TH1F* h1 = new TH1F("h1","h1",5,0,5);
> for(int i=0;i<5;i++) {
> h1->SetBinContent(i,i*100.);
> }
> TH1F* h2=h1->Clone();
> h2->SetNameTitle("h2","h2");
> h1->SetLineColor(3);
> h1->SetFillColor(3);
> h2->SetLineColor(2);
> h2->SetFillColor(2);
> h1->Sumw2();
> h2->Sumw2();
> //////////////////////////////////////////////////
> //<Performe multiplication>
> // fuction * constant
> // 1) 1 * 0.1
> // 2) 0.1 * 1.0
> TF1* f1 = new TF1("f1","1.0",0,5);
> TF1* f2 = new TF1("f2","0.1",0,5);
> h1->Multiply(f1,0.1);
> h2->Multiply(f2,1.0);
> //////////////////////////////////////////////////
> //<Draw hist and show bin error>
> h1->Draw("E2");
> h2->Draw("E same");
> cout << h1->GetBinError(1) << endl;
> cout << h2->GetBinError(1) << endl;
> }
>
> Best Regards,
> Hajime
>
> Hajime Nanjo
> International Center for Elementary Particle Physics,
> University of Tokyo
> E-mail: nanjyo@icepp.s.u-tokyo.ac.jp
This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:09 MET