Apr 10/25 PionLT/KaonLT Analysis Meeting Notes
                 ----------------------------------------------
                                (Notes by GH)

                     Today: KaonLT will be discussed first

Present
-------
Regina - Garth Huber, Nacer Hamdi, Alicia Postuma, Muhammad Junaid,
   Nathan Heinrich, Vijay Kumar, Ali Usman
JLab - Dave Gaskell
Virginia - Richard Trotta
FIU - Pete Markowitz
CUA - Tanja Horn

Ali
---
pi+Delta0 BSA t-phi binning
- revised binning for Q2=3.0, W=3.14 high statistics setting
  - change from 2 to 3 t-bins, upper limit of good range remains 0.70
  - bin limits: 0-0.33, 0.33-0.42, 0.42-0.70
  - excluded region: 0.70-1.10
- check phi coerage for these bins
  - MM plots for each t-phi bin L,C,R SHMS settings look promising

- Q2-3.0, W-2.32 setting
  - change from 1 to 2 t-bins, upper limit remains 0.90
  - bin limits: 0-0.6, 0.6-0.9
  - excluded region: 0.9-1.10
  - MM plots for each t-phi bin will have some challenges but Ali is confident
  - definitely a drop in statistics from 1 bin, but good phi coverage

- Nacer asks some questions about how the background near the Delta peak is
  handled
  - full MM region histo including pi+n peak is used for Background fit
    - pi+n MC is fit to neutron peak
    - SIDIS MC is fit to high MM region
  - after the Bkd is fit, use zoomed in MM region for piDelta MC fit
  - piDelta MC peak shape is compared to data after Bkd subtraction for
    consistency

- 2 higher Q2 settings will not be rebinned due to low statistics
- Ali hopes to show Bkd subtracted fits for all 3 revised settings next week
- while fitting MCs to data, will do some systematics checks in parallel (fit
  dependence)

Richard
-------
KaonLT high Q2 LT-separations
- Q2=3.0, W=2.32 (same setting as Ali just showed, but K+ instead of pi+)
  - had some difficulty getting 0th iteration to work for this setting
  - ended up trying Q2=4.4 functional form instead of Q2=3.0, W=3.14 form and
    it seems to work a bit better, now getting iterations to converge
  - result is a huge difference in R=Data/MC ratio between high and low epsilon
  - low epsilon R about 1 for t=0.40-0.75 bin
  - high epsilkon R about 2 for same bin
  - large change in ratio seems to be due to large change in SIMC yields
    between high and low epsilon, while data yields are more consistent
  - this seems to imply a significant change in L/T ratio between this setting
    and Q2=4.4

Vijay
-----
PionLT Q2=0.425 LT-separations
- updated last t-bin edge to 0.60 from higher value discussed last week
  - revised t-phi plots shown for low, mid, high epsilon
- working on MM offset for all settings, i.e. match pi+n peak of data to MC
  - old analysis script is breaking, it did too many tasks at once
  - new script will be simpler
  - plans to show MM plots next week

Nacer
-----
KaonLT Q2=0.50 LT-separations
- working on extracting yields per t-phi bin for Data and MC
  - scripts are done, checked on some data runs
- checking error propagation of ROOT
  - confirmed that ROOT is propagating the statistical errors of the data
    correctly, but it appears ROOT has some issues when the normalization
    itself has an uncertainty
    - i.e. Y=Counts/Qeff, the uncertainty in Qeff is not propagated correctly
  - adding this uncertainty "by hand" in the code, code is checked and should
    be correct now
- will meet with Richard this week on updating weights each iteration w/o
  rerunning SIMC

Alicia
------
Missing tracks Geant4 simulation
- finished 1st draft of TechNote and got code running on the group's server
  (lark), that part is relevant for improving the transportability of the code
  with slightly different G4 versions
- the TechNote will be sent shortly to Regina group and Richard, who should
  check if the note provides good instructions for calculating the correction
  for different analyses and that the code works correctly for them
- started calculating missing tracks correction for KaonLT u-channel
  p(e,e'p)omega data

- still no update on BSA paper from PLB, hopefully soon

Nathan
------
Lumi study coincidence blocking correction
- Dave will try to look at Nathan's CoinTime plots this weekend and get back to
  him on this

Junaid
------
PionLT Q2=3.85, W=2.62 LT-separations
- will change from 7 to 5 t-bins, as we discused last week and then calculate
  yields
  - yield calculation is different than Nacer's
  - Junaid's version uses separate scripts for each part, while Nacer's is a
    big all-in-one code (see Vijay's experience above)


Next Meeting
-------------
- Thur Apr 17 @ 15:30 Eastern/13:30 Regina
  - PionLT will go first
  ** NOTE THE NEW TIME **