Re: [ROOT] TMultiGraph fit gives me a bad chi square.

From: Venkatesh Kaushik (venkat@hepmail.uta.edu)
Date: Mon Nov 17 2003 - 17:52:35 MET


Hi,

The values on the X- axis are in GeV, whereas the ones on the
Y-axis are in MeV, so i multiplied the errors and mean-values
on Y-axis by 0.001. to make both the axes be in same units. Is that
a mistake?
Venkat

On Mon, 17 Nov 2003, Rene Brun wrote:

> Hi,
>
> Your errors cannot be right. They are far too small. This makes the chisquare
> too big.
> Why do you multiply the errors by 0.001 ?
>
> Rene Brun
>
> Venkatesh Kaushik wrote:
> >
> >  Hi ROOTers,
> >  I have two sets of data points, which I am supposed to fit straights line
> >  to. I try to fit both these with pol1, I observe two things
> >  1) the chi square / dof value is horrible.
> >  2) visually, i don't see the fit has done a good job.
> >
> >  i am not sure what's happening. are there any default fit
> >  strategies/parameters that is messing up my fit?
> >
> > I have attached my code and the program output for your kind perusal.
> > Thanks for the help.
> > Venkat.
> >
> >   --------------------------------------------------------------------------------
> >                          Name: Eliveonly_ecal.gif
> >    Eliveonly_ecal.gif    Type: GIF Image (IMAGE/gif)
> >                      Encoding: BASE64
> >
> >                   Name: Ecal_live.C
> >    Ecal_live.C    Type: Plain Text (TEXT/PLAIN)
> >               Encoding: BASE64
>



This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:16 MET