RE: Rebin

From: Manuel Mussini <Manuel.Mussini_at_bo.infn.it>
Date: Tue, 27 Nov 2007 09:04:19 +0100


But this is a simple mean!
The weight of the "old" bins is not considered!

Bye...

> -----Original Message-----
> From: owner-roottalk_at_lxbuild091.cern.ch [mailto:owner-
> roottalk_at_lxbuild091.cern.ch] On Behalf Of Edmond Offermann
> Sent: Tuesday, November 27, 2007 02:48
> To: Manuel Mussini; roottalk_at_lxbuild091.cern.ch
> Subject: Re: [ROOT] Rebin
>
> Fill another histogram with 1 in every bin: TH1::Fill(x, 1.) .
> Rebin this histogram in the same way . Now divide your
> original rebinned histogram through this rebinned histogram.
>
> Eddy
>
> --- Manuel Mussini <Manuel.Mussini_at_bo.infn.it> wrote:
>
> > Hi,
> > I have a TH1D filled with TH1::Fill(x, 1./x)!
> > At "high" x there is a low statistics and errors are big so I'd like
> > to rebin
> > the histo increasing the bin width progressively. Using
> > TH1::Rebin(Int_t, const
> > char*, const Double_t*) I can merge the bins as I like but the histo
> > is no more
> > "normalized" (there are steps each time the width increases)!
> >
> > Any idea on how to fix this issue?
> >
> > Bye...
> >
> >
Received on Tue Nov 27 2007 - 09:04:50 CET

This archive was generated by hypermail 2.2.0 : Tue Nov 27 2007 - 11:50:01 CET