Dear ROOT people,
I am new in the forum but I have been using ROOT a little for my work. Here is my question (don’t know if I am posting in the right place):
I am fitting a graph using the basic fitting routines and using TVirtualFitter to retrieve the fit results and I can obtain the uncertainty on the parameter value on both sides using the code below (I have 4 parameters):
graph->Fit(f1,"E");
TVirtualFitter *fitter = TVirtualFitter::GetFitter();
double eup,elo,epar,glob;
for (int i=0;i<4;i++)
{
cout << "Paramter "<<i<<": "<<fitter->GetParameter(i)<<"\t"<<fitter->GetParError(i)<<endl;
fitter->GetErrors(i,eup,elo,epar,glob);
cout << "par "<<i<<": " <<eup <<"\t"<<elo<<"\t"<<epar<<"\n";
}
Because of poor statistics of the data, for one of the parameters I obtain that the lower error is 0.
Paramter 0: 5.43812e-06 5.46689e-06
par 0: 1.60752e-05 0 5.46689e-06
I was wondering if it is possible to produce the parameter profile to test if there are some problems in the fit.
What I would like to have is the value of the chi2 when I scan the possible values of the parameter.
I have been looking into the documentation but I could not find what I need. Is there some easy method?
I could write my code to do this job, but if there is already a method implemented it would be very helpful.
Cheers!