Re: [ROOT] 1 d histogram fitting

From: Edmond Offermann (edmondoffermann@yahoo.com)
Date: Wed Sep 10 2003 - 02:40:52 MEST


Hi,

The number to observe is the ratio between
the data and the integral of the fitted Gaussian.
You would like this number to be close to 1.
A good starting point is to choose binning
in your histogram such that sigma >= 1.5 bins .

The value of chisq/ndf has an expectation value of
1 but is not a good indicator for bin size.
Deviations from 1 are more likely from the data
not following Gaussain statistics.

Eddy


--- Venkatesh Kaushik <venkat@hepmail.uta.edu> wrote:
> Hi ROOTers,
> i have a simple question. I have a set of data which
> i know is nearly
> gaussian, and i would like to plot the histogram.
> what i don't know is how am i going to decide on the
> number of bins and
> the limits for defining a TH1F histogram object?
> 
> i know if you change the number of bins the mean
> value and the width of
> gaussian varies. i need to be able to get a good
> mean value and a good
> value close to 1 for chisq/ndf if i fit a gaussian.
> any comments or
> readily available code?
> 



This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:15 MET