Update 09.18.2018 -- Run 9 pp: Default Unfolding vs. Efficiency +- 5% (R = 0.2 Charged Jets)
Following up from yesterday:
It would be better to compare apples-to-apples, i.e. use the same 'eTtrg' range in both the parameterized Pythia response and the embedding response. Below compares the unfolded solutions for both the embedding and Pythia responses (with +- 5% on the tracking efficiency). In each bin of 'eTtrg', the embedding and Pythia responses have the same 'eTtrg' range.
And strangely, the agreement seems worse (especially in the 'eTtrg = 9 - 11' bin)... So here's the how the efficiencies compare:
However, I also calculated the jet-matching efficiency (+-5% on the tracking efficiency) in Pythia for 'eTtrg = 9 - 20' GeV, the same trigger-range as I originally used in the embedding. The agreement between the unfolded solutions is substantially better (though there are isolated bins where both Pythia solutions are above or below the default). However, the agreement is about the same as using the 9 - 20 (embedding) response for the default vs. the 9 - 11 and 11 - 15 (Pythia) responses for the systematic variations.
And here are the efficiencies:
- dmawxc's blog
- Login or register to post comments