[Rivet] histograms in Rivet

Hendrik Hoeth hendrik.hoeth at cern.ch
Fri Aug 31 00:32:58 BST 2012


Hi Hannes,

> I thought, if I do not call the "normalize" procedure at the end, the
> histos would not be normalized to bin widths, but it seems they still
> are. Is this correct ?

That's correct. This is why we call it a histogram, not a bar chart,
since histograms are invariant under changes of the binning. I know that
this concept hasn't reached the ROOT developers yet ...

> And if yes, is there a easy way to obtain a histo without
> normalization to bin width ?

Either fill the histogram with the bin width as additional weight (e.g.
for histograms with variable binning) or scale the histogram in
finalize() with the bin width (in case of equal bin sizes).

Cheers,

    Hendrik

-- 
Journalist: "Can you explain 5 sigma as 99.999something percent?"
Fabiola G.: "Sure, that's 3*10^-7."
(after the CERN Higgs seminar, 4 July 2012)


More information about the Rivet mailing list