Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

combine_limit example showcases bad practice? #97

Open
andresti opened this issue May 27, 2019 · 1 comment
Open

combine_limit example showcases bad practice? #97

andresti opened this issue May 27, 2019 · 1 comment

Comments

@andresti
Copy link

In the combine_limit example both 1-sigma and 2-sigma uncertainties are added to the expected value, but this seems to go against the CMS Higgs HepData recommendation.
From https://twiki.cern.ch/twiki/bin/viewauth/CMS/HiggsHepData:

For expected limits ("Brazil plots"), please note that it is not recommended to give the 1σ and 2σ bands as independent errors, since the "sum errors" button will sum them in quadrature and give meaningless results. It is better to give the expected limit twice (as two columns/graphs), one with the 1-sigma and the other with the 2-sigma uncertainty.

@AndreasAlbert
Copy link
Collaborator

Hi @andresti,
I am not sure I would call this 'bad practice', since the error summing is just a result of the hepdata website visual interface, which is not set in stone and should not be taken for granted. For someone who wants to re-use your data, having 1 and 2 sigma uncertainties on one variable poses absolutely no technical issues. Therefore, I would argue that this is more a question of personal taste at this point.
That being said, I think we (as a community) should come up with actual recommendations on how to handle this and similar "style issues" so that everyone uses compatible conventions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants