Document how to make the "SUSY standard" upper limit table

[kratsg]

Summary

HistFitter has code to make an upper limit table plus a separate call for $p_0$. This is related to #2262 and should produce a table like below:

We should document how to do this in pyhf as a learn notebook or similar.

Documentation Page Link

https://scikit-hep.org/pyhf/examples.html

Code of Conduct

• [X] I agree to follow the Code of Conduct

[kratsg]

See also @longjon929's slides about this table.

[kratsg]

2L2J is a good analysis to use for comparing to what HistFitter made.

• Table 20/23 are UL tables: https://arxiv.org/abs/2204.13072
• HEPData record should have HiFa JSONs: https://www.hepdata.net/record/116034

[kratsg]

Pivoting to EWK Compressed as those workspaces are somewhat less convoluted to deal with and their tables are also from HistFitter

• http://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/SUSY-2018-16/
• https://www.hepdata.net/record/ins1767649

See table 10 for example: http://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/SUSY-2018-16/tab_10.png

UPDATE: these are the multi-binned workspaces, and not the single-binned discovery workspaces that ended up serialized.

[kratsg]

Some references:

• https://twiki.cern.ch/twiki/pub/AtlasProtected/StatisticsTools/Frequentist_Limit_Recommendation.pdf
• https://cds.cern.ch/record/1955204/files/CERN-THESIS-2013-335.pdf (page 59-60 of Sarah Williams thesis)
• https://indico.cern.ch/event/1145930/contributions/4809767/attachments/2430295/4167778/22-04-28-susy-stats.pdf