Update 09.17.2018 -- Run 9 pp: Unfolding Systematics, Pi0 (R = 0.2, 9 - 11 and 11 - 15 GeV)
The plots below show the systematic uncertainties on two of the unfolded pi0 spectra, R = 0.2 charged jets for 'eTtrg = 9 - 11' GeV and 'eTtrg = 11 - 15 GeV'. The legends indicate the different systematic variations and the unfolding method, regularization parameter, and prior used in unfolding each variation. Note that the Pythia6 prior is triggered on charged hadrons, and the Pythia8 prior is triggered on pi0's.
The default unfolding for the two trigger ranges are:
- Bayesian algorithm, k = 2 (9 - 11 GeV) and k = 3 (11 - 15 GeV) respectively
- Pythia6 prior (R = 0.2 charged recoil jets triggered on charged hadrons with 'eTtrg = 9 - 20' GeV).
Using 'eTtrg = 9 - 20' GeV vs. 'eTtrg = 9 - 11' or 11 - 15 GeV results in a negligible difference. The different variations assess the following systematics:
- method 1: using a different unfolding method (SVD)
- method 2: using a different "unfolding" method (bin-by-bin corrections)
- k + 1: using a different regularization parameter (set to be the default value + 1)
- prior 1: using a different prior (pi0-triggered Pythia8 spectrum)
- prior 2: using a different prior (levy distribution)
- prior 3: using a different prior (exponential distribution, 'f(x) = exp[x / tau]')
- eff - 5: tracking efficiency lowered by a relative 5%
- eff + 5: tracking efficiency raised by a relative 5%
- res. 1: using a different parameterization of the tracking resolution (variation 1, see below)
- res. 2: using a different parameterization of the tracking resolution (variation 2, see below)
The default parameterization of the tracking resolution and its two variations are as follows.
- default: sigma(pT) = 0.0142 + (0.0107 * pT) + (0.0013 * pT^2)
- variation 1: sigma(pT) = 0.0445 + (0.0070 * pT) + (0.0013 * pT^2)
- variation 2: sigma(pT) = -0.026 + (0.020 * pT) + (0.003 * pT^2)
They're described in more detail in the post below.
https://drupal.star.bnl.gov/STAR/blog/dmawxc/update-05072018-run-9-pp-resolution-study-final-parameterizations
As always, the relevant ROOT files are attached.
Update [09.17.2018]: added the same plots but with linear y-axes on the ratios.
Update [09.17.2018]: a more detailed comparison of the default unfolding to the efficiency variations can be found in the link below.
- dmawxc's blog
- Login or register to post comments