Hi Catalin, It is difficult to come with a diagnostic without looking at your code. Could you send me - a small root file containing your 2 histograms - a short macro (or program) that I can run myself Rene Brun catutza@mps.ohio-state.edu wrote: > > Hi Rooters, > > I have a question regarding TMinuit. > > I have two Monte Carlo histograms (signal and background), and I am > generating a fake data histogram by drawing N1 'events' out of the signal > histogram, and N2 from the background histogram. > I use then TMinuit to fit the fake data as the sum of the two MC histos - > to see if I get numbers close to N1 and N2. > > I do this many times; the fit always returns about the right value for N1 > and N2. However, the error returned by the fit appears to be the same > every time (it slightly varies within a very small range). > > Has anybody seen this (or something similar) before? Is there a bug in the > way the errors on the parameters are computed? > I am using ROOT Version 2.22/06. > > Thanks for the help, > > Catalin Ciobanu, CDF > e-mail: catutza@mps.ohio-state.edu
This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:29 MET