Questa è una vecchia versione del documento!

http://cms.cern.ch/cds/SMP-15-009

Stefano:

ce l'ho messa tutta, ma 'sto malloppo e' veramente pesante, non sono riuscito a leggerlo tutto con attenzione.

Metto quei pochi commenti che ho. Non so se avrebbe senso dire “The paper is very dense, very technical and quite detailed, we suggest to streamline it a bit in order to make it more enjoyable”

Section “Data and simulated samples” This section is very long and technical and does not attract the reader into getting further into the paper. For the non-expert it is hard to understand the reasons behind the choices of generators, tunes and PDFs. For example it is not clear why at least three different PDF sets are used. I would recommend moving this section down further in the paper. I would add in its stead a section which would give a quick outline of the analysis strategy.

L 256: please clarify right away if the sample consists of data or simulation

The “Analysis strategy” section appears only at page 10. I wonder if section 6 could be renamed “Yield extraction”, and if a new “Analysis Strategy” section could be added around page 4, as already suggested. This would bring the reader faster into the philosophy of the paper.

L329 “based on fits to distributions ..” but then the nature of these fits is elucidated only more than 50 lines further.

L379 “a chi2-test fit” : probably the word “test” is in excess ? It is not clear which distributions exactly are fitted to extract the yields, or if it is a combination that is used. What is the fitting function ? Is it a template fit in which the simulated shapes are fitted to data (we know the answer is yes from browsing through the analysis note, but it should explained clearly here) ? How exactly is the uncertainty on yields calculated ? Are there systematic uncertainties on the yields ?

Figure 4: hatches almost invisible