Workbook chapters

[vincecr0ft]

The picture I have in my mind is to separate the workbook into larger topics each with various notebooks on the various contained topics.

• Likelihoods: profiles, intervals and systematics. Using standard RooStats tools to extract useful information from the likelihood of a model or series of models.
• Models: channels, shapes and normalisations. Construction of Histfactory objects and exporting workspaces. Definition of Histfactory nomenclature (Channel, Region, etc). Explanation of interpolation/morphing scheme.
• Hypothesis testing: deconstructing and explaining the standard runSig.C and AsymptoticsCLs.C macros. Presenting results.

I understand that there will be some overlap between these. But if a small amount of repetition can be coupled with a style that facilitates people skimming over each chapter in order to understand the particular issue that they're dealing with.

[vincecr0ft]

I'm currently starting work on the Models section in order to build an example workspace that can be used for the other sections.

I'd love some guidance as so what kind of systematics etc I should include (such that they can provide useful cases for the various other sections)

[vincecr0ft]

As agreed with StatsForum Conveeners, structure should be rich text documentation in chapters with links to images produced by notebooks that can be checked out in swan or loaded in binder. These should fall into 4 sections.

• [x] Statistics.
• [ ] Tool reference usage
• [ ] Recommendations for HEP analyses
• [ ] Examples