Nuno questions 290715

Dear Nuno, Many thanks for your questions and comments. Here are some answers (we are working in the others!).

1) selection.would you mention the procedure and criteria utilised in selection optimisation.

Answer: In the case of the Bs->f0(980) channel the selection criteria follows what was made for the Branching Ratio work (AN2012_363), for Bd->K* we used the same cuts because we wanted to follow the same procedure as in the f0 to have this as our crosscheck channel. In the Lambda_b and Bd channels (Xib used basically the same cuts) the selection used is like in previous work about Lambda_b lifetime (AN-14-045) by Frank.

2) physics backgrounds: partially and mis reconstructed b decays.a detailed description of these seems to be missing. please include MC plots showing the various components affecting each final state. presumably, MC-derived shapes are employed in the PDFs.

Answer: Unfortunately there is not official MC to do these studies. However we have only found significant background (of this kind) in the channel Bs->f0(980) where the technique used is described in Appendix E of the analysis note.

3) cabibbo suppressed components: this seems to be missing for example for the B+ case. The JpsiPi bump should sit tothe right and under the main peak.

Answer: Work in process

4) long lived B+ background: this was discussed at the earlier stages if the review. please point to theconclusion of the studies performed. (cf fig.3, tab.2)

Answer: After several studies with different vertices and finding that all of them have these long components (looking in the sidebands region), the decision was to model it phenomenologically, ie using exponential components (as many as necessary).

5) K* swap: you make a selection. in some fraction of the times, an incorrect mass assignment is made. this results in a mass assignment that differs from (is broader than) thesignal. a term needs to be added to the likelihood to describe these candidates. The lifetime parameter would still be that of the signal. this is besides the systematic considerations in point 6 of section 6.

Answer: Work in process

6) ct efficiency: the ct distribution may be biased by online as well offline selection andreconstruction. were effects from the latter two considered?a data-based calibration would be suitable. use for example the 2011 B+ dataset(that is otherwise not utilised in the analysis). or the 2012 unbiased sample fromparked data. attempt to extract the efficiency parameters from a fit to this sample.

the linear model in eqn.26 may be appropriate; others could be also tried. then, inthe final nominal fit used to extract the lifetime, use a ct-efficiency functionwith these parameters constrained (via gaussian constraints aka penalty functions).

Answer: Work in process

7) ct resolution: please use the high-statistics trigger-unbiased parked-data 2012 sample, by fittingthe prompt components, to calibrate the ct resolution. there were other extensivestudies of ct resolution performed in the connect of the Bs lifetime difference andphi_s paper (BPH-13-012). please attempt to propagate these studies as cross checksof related effects.

Answer: yes, we will make this studies as soon as Sara sends the unbias sample, and will review the reference you mention to see what could be done in our case.

8) primary vertex selection: l.104-7 comment: this issue is being completed in the next version of the AN

Answer:Y es, now it is present in the latest version of the note

9) notation: for consistency with other related CMS publications, it would be nice to use ctinstead of lambda and of PDL. the is relevant for the PAS/paper.

Answer: Ok. we will change the notation in the next update of the note.

-- JhovannyMejia - 2015-07-30

Edit | Attach | Watch | Print version | History: r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r3 - 2015-07-31 - JhovannyMejia
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback