- marr's home page
- Posts
- 2023
- 2022
- 2021
- May (1)
- 2020
- 2019
- 2018
- November (1)
- October (1)
- September (1)
- August (1)
- May (2)
- April (5)
- March (1)
- February (1)
- January (1)
- 2017
- December (1)
- November (1)
- October (3)
- September (1)
- August (2)
- July (1)
- June (3)
- April (3)
- March (2)
- February (1)
- January (5)
- 2016
- 2015
- 2014
- My blog
- Post new blog entry
- All blogs
HF-PWG minutes 2017/06/01
1) Systematic uncertainty for Run12 pp Jpsi polarization - Siwei Luo
a. pT smearing: suggest to take a look at what Bingchu did for the J/psi cross-section measurement using the same data set, and use the same pT smearing factor and formula as the default for this analysis. The pT smearing factor should come with an uncertainty, which needs to be propagated. One can vary the pT smearing within its uncertainty, and take the average of absolute differences to the default value as the uncertainty.
b. Weight function: take Bingchu's spectrum and do the fitting. The resulting uncertainty on the fit parameters can be used to calculate the uncertainty on the polarization parameter.
c. NHitsFit: currently, NHitsFit is varied by +/- 2, which seems a bit small since the change in the efficiency is very small (slides 23-26). Suggest to look at previous HF and LFS analyses to see what are the typical changes they do.
d. nSigmaE: show the fitting to upper and lower limits, where statistical error bars also need to be taken into account.
e. DsmAdc cut: need to apply adc0 cut consistently in data and embedding. Need to check why the change in efficiency is visible on slide 35, but not so much on slide 36. Please list DsmAdc and Adc0 cut values. Suggest to change DsmAdc by +/-1 instead of +/-5%.
f. 1/beta: check the tails of the 1/beta distributions to see whether a Gaussian fit is justified.
g. Add figures to show the systematic uncertainty for each source as a function of pT
h. Need to estimate the uncertainties for NHitsDedx, dca, etc.
i. For all the plots with open and solid markers, please increase the size of the markers and use different colors.
- marr's blog
- Login or register to post comments