[Rivet] Including weighting data in analysis?

[Andy Buckley]

On 08/08/13 18:26, Tim Martin wrote:
> Dear Rivet,
> 
> We are considering how best to write a rivet class for a analysis
> which requires access to a collection of data weighting histograms
> which would be used in lieu of a hadron level definition. We
> understand that this is a non-standard way of doing things, but looks
> to work well for this specific analysis.
> 
> Is there any precedence for this, or anything more fancy than a bunch
> of "float histogramData[] = {0.1, 0.2....};" we should look to use?

Hi Tim,

Apologies for hugely delayed response! By weighting information do you
mean effectively a "forward folding" factor so that particle-level MC
can be compared to uncorrected/semi-corrected data? The only example of
that so far is a very simple "lossy final state" processing that we used
for the first CDF underlying event paper -- it just randomly throws away
8% of particles for comparison to the published data!

Anyway, you know that we prefer fully corrected data, but there's no
reason not to do it this way if you're happy about the correction
factors (and the data errors include the uncertainty that they imply?)
But there's no specific mechanism for it: double[] would be just fine :-)

Andy

-- 
Dr Andy Buckley, Royal Society University Research Fellow
Particle Physics Expt Group, University of Glasgow / PH Dept, CERN