Greetings Rootmeisters,
It seems that log-likelihood fitting is integerizing bin
contents before fitting. Is this intended behavior?
The following code snippet demonstrates the problem:
if the histogram is scaled by a factor of 10, the fit is
done correctly; otherwise, it fails to converge in MIGRAD.
This is ROOT 2.23/12.
{
TH1F* hist = new TH1F("hist", "hist", 5, 0, 5);
TF1* func = new TF1("func", "[0] + [1]*x", 0,5);
hist->Fill(.5,.1); hist->Fill(1.5,.2); hist->Fill(2.5,.3);
hist->Fill(3.5,.4); hist->Fill(4.5,.5);
// hist->Scale(10.);
hist->Fit("func", "VL");
}
This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:24 MET