Re: [ROOT] TF1::Draw() with error bars

From: icpark@mail.cern.ch
Date: Thu Nov 15 2001 - 16:30:18 MET

Next message: Roberto Catanuto: "[ROOT] PAW locate"
Previous message: Isidro.Gonzalez.Caballero@cern.ch: "Re: [ROOT] CERN RH7.2 and ROOT"
In reply to: Rene Brun: "Re: [ROOT] TF1::Draw() with error bars"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi Rene,

I just used profile->Fit("pol1"), just like Fit("gaus"). 

My understanding is : 

- profile histogram returns error as "spread/sqrt(N)", where "spread" is
the width of the distribution of "y" at bin "x". 

- prof->Fit("pol1"), fit a straight line using chi2 method, ie, minimize
the sum of deviations, by MINUIT. In this case, ie, straight line fit with
errors, it's similar to Least Square fit. For LSF, error calculation is
easy.

Emm... I can send a macro if you wish to try.

Thank you,
Inkyu

Next message: Roberto Catanuto: "[ROOT] PAW locate"
Previous message: Isidro.Gonzalez.Caballero@cern.ch: "Re: [ROOT] CERN RH7.2 and ROOT"
In reply to: Rene Brun: "Re: [ROOT] TF1::Draw() with error bars"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:51:08 MET