Hi Pascal, This problem had already been reported by Marco van Leeuwen and fixed in version 2.24/02. See release notes. (here is the extract) - Many thanks to Marco van Leeuwen for reporting a problem with TGraph::Fit when fitting with a polynomial. The problem has been fixed in TGraph::LeastSquareFit. The last data point was ignored by this function computing the initial value of parameters. Rene Brun On Tue, 11 Apr 2000, Pascal Perrodo wrote: > > Hi rooters, > > I have problem doing fits using Minuit in Root. You can find attached in > this mail two examples where a simple polynomial is fitted to points into > a TGraphErrors plot. > > The problem is that this fits don't converge all the time even with a low > degree polynomials (degree 2). > > Do you see any explanation ? > > Thanks in advance. > Cheers. > Pascal Perrodo. > > +-------------------------------------------------------------------------+ > | Tel : 04 50 09 16 32 (from France), 00 33 4 50 09 16 32 (from abroad) | > | FAX : 04 50 27 94 95 (from France), 00 33 4 50 27 94 95 (from abroad) | > | | > | LAPP, Chemin de Bellevue, BP 110, F-74941 Annecy-le-Vieux CEDEX | > +-------------------------------------------------------------------------+ >
This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:23 MET