hello
I want to fit a TGraph which have a unsymmetrical gaus distributing. This TGraph have the same “constant” and “mean”, but have different “sigma” less than “mean” and more than “mean”.
Would you help me? Thank you!!
foolfish
6.8 [/b]
see example below
root > .x DoubleGauss.C
Rene
Double_t mydgauss(Double_t *x, Double_t *par) {
Double_t arg1 =(x[0]-par[1])/par[2];
Double_t arg2 =(x[0]-par[1])/par[3];
Double_t r1 = TMath::Exp(-arg1*arg1);
Double_t r2 = TMath::Exp(-arg2*arg2);
return par[0]*(r1+r2);
}
void DoubleGauss() {
TF1 *f1 = new TF1("f1",mydgauss,-3,5,4);
f1->SetParameters(1,0,1,2);
TH1F *h1 = new TH1F("h1","double gauss",100,-3,5);
h1->FillRandom("f1",10000);
h1->Fit(f1);
h1->Draw();
}
Thank you very much!!
foolfish