Setting limit of Chi^2 & Fit results box on TGraph2D

ROOT
1

I have problems as following

  1. I am trying to fit some TF2 function and i cannot find on manuals how to change desired Chisquare (should it be function: TF2::Setchisquare(double) ?) to improve my fits since fitting procedure ends with something not nice at all
  2. Is there any easy way to plot fitted parameters on TGraph2D ? with TF1 it is working nicely with TGraph and gStyle->SetOptFit(char *) but the same function does not work at all with TGraph2D.
2

I already replied to your mail to root-bugs. Why do you resubmit it here?
My answer was:

  1. I am trying to fit some TF2 function and i cannot find on manuals how to
    change desired Chisquare (should it be function: TF2::Setchisquare(double) ?)
    to improve my fits since fitting procedure ends with something not nice at all

Use TF1::setChisQuare (a TF2 is a TF1)

  1. Is there any easy way to plot fitted parameters on TGraph2D ? with TF1 it
    is working nicely with TGraph and gStyle->SetOptFit(char *) but the same
    function does not work at all with TGraph2D.
    see tutorials graph2dfit.C or graph2derrorsfit.C

Rene

3

I get confused about forum and all that stuff and thought that this open forum could help. Nevertheless function:
TF1::SetChisquare(double) does not work and fit is finishing like without using this function.
And function which you wrote: TF1::SetChisQuare does NOT exist in class TF1. So i still looking for something working by searching through ROOT manuals, hunting for bugs in my code, etc.
Best regards

slawo

4

The correct spelling is simplyTF1::SetChisquare
Cheers,
Philippe.

5

Yes i know. But the problem is that it is not working like i think it should.
So i do:
TF2 *aaa = new TF2(…)

aaa->SetChisquare(2) ;

and fit stops anyway around Chi^2=1400

So i get confused what is wrong. Nevertheless i will play around with it i coming days since i localized also huge mistake in derivation of my fitting function. I am trying to do several things in the same time so sorry for bother you. Anyway new version of formula improved a lot (chi^2 from 2900 down to 1300 so i made progress). But for my understanding of root it is still bad :wink: I need to spend a little bit more time with it. But it is quite nice tool :slight_smile: And as one of my colleagues said: “root’s manuals has some impurities but try to find one on at least the same level”. I appriciate your interest.
best regards

slawo