[Color_transp] [EXTERNAL] Weighted efficiencies

[John Matter]

Hi all,

I've been using the TEfficiency class to calculate the uncertainty for
binned efficiencies.
https://urldefense.proofpoint.com/v2/url?u=https-3A__root.cern.ch_doc_master_classTEfficiency.html&d=DwIBaQ&c=CJqEzB1piLOyyvZjb8YUQw&r=XtVl8BY42jbbWh0Vu1ygqgFBYEJMENirtKAh3P_9ayw&m=p_JDxPoCy4gZfVxXIuO-KL2JBeXRVEG5BT3OinPWKoc&s=KxD1ajQNJhhzlM5UEug-Gl5BtFYBw3o6bzeDAT4lzZ4&e= 

I thought the default setting used the Bayesian method I've mentioned in
the past, outlined in the link below.
https://urldefense.proofpoint.com/v2/url?u=https-3A__lss.fnal.gov_archive_test-2Dtm_2000_fermilab-2Dtm-2D2286-2Dcd.pdf&d=DwIBaQ&c=CJqEzB1piLOyyvZjb8YUQw&r=XtVl8BY42jbbWh0Vu1ygqgFBYEJMENirtKAh3P_9ayw&m=p_JDxPoCy4gZfVxXIuO-KL2JBeXRVEG5BT3OinPWKoc&s=BaMx6yf-0-pLJfCMCF-ZLC1iRWiJ-ZvmjqhbThHhd7U&e= 

However, it turns out the default method that I've been using
is Clopper-Pearson. The options for confidence intervals are described in
section IV of the ROOT documentation above.

I'll upload my weighted efficiencies and unweighted efficiencies to the
spreadsheet. Changing the binned efficiencies' uncertainty calculation only
requires changing one line of code. That'll change the weighted values, but
I have a pipeline setup that output a CSV that can be easily uploaded to
the spreadsheet.

- John
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/color_transp/attachments/20200806/95c19ae3/attachment.html>