Difference: AnalysisTaskListOct22 (1 vs. 2)

Revision 22011-04-21 - PeterWinter

Line: 1 to 1
Changed:
<
<
-- BrendanKiburg - 22 Oct 2008

The idea here is to list open questions, outline the proposed method for answering the questions, and prioritize the tasks.

Questions Generated from recent teleconference

Addressed during week of 10/15/2008

  • Double check the (y,z) distribution with nContEH ; also look at the y distribution where each z slice is normalized, to see if things are truly downward biased
    • Interesting results to both questions in this elog
  • I need to redo the shrubs where the z_electron is the z_tracked back to the tpc frame
    • First results for 1 file: elog
  • Response to Peter's elog estimating the lifetime deviation due to scatters: elog
  • Peter W. asks for cvs update with new shrub additions (best entrance flag, HasBestEntranceIgnoringSparks(), etc) : updated 10/15 , afternoon

Remain Open

  • Dave suggets start time scan for midplane gondolas, without nContEH cut. If OEH mustops leak into 1EH mustops, this should disappear at some point and become stable.
  • Peter Winter suggests a similar stop time scan for downward going ones
  • I suggest adding up all upward going guys (1-3, 16->14) and looking for gondola variations

  • For sara: Given a small abs(theta), look to see if the nCont distribution is different for different y-slices

  • Use the event display to examine nExEL >= 16. Are these scatters, sparks, something else?

  • Repeat the nExEL scan with a cut on the fit to the extra EL pointing/not pointing to the stop
    • This involves setting up some simple 2nd-level tracking to fit the extra pixels
    • Check auxptr and TTwoLine info in the mustop tree (for visualization on event display and observe available chi-square information)
  • Check hypothesis that the decay modifies the nContEH part of the track by making requirements that the Michel is upward going
  • Dave suggests that making plots versus theta (or gondola z), rather than just phi
  • Compare the muplus fits to the muminus
    • i peter refines the 7-parameter fit
    • make use of the par-scan macros
  • make simple, spatial plots (x,y,z vs, nExEL, nContEH)
    • run shrub level
    • add to the main mta analysis
  • Define the nExEL variable for nIslands > 1 (currently no extra EL pixel information is collected)
    • NExEL for island i = (nExEL - sum EL in all real islands)
    • Consider using IsNotAlone for cuts
  • Peter W. asks for cvs update with new shrub additions (best entrance flag, HasBestEntranceIgnoringSparks(), etc)
  • Simple 1 parameter model to show how distortion for a given gondola for n nContEH depends on the relative statistics in bin n and n-1
  • gond:dt:nContEH | angle of decay electron (upstream vs downstream)
  • gond:dt:nContEH | angle of decay elec | y_stop height

Addressed during week of 10/8/2008

  • make simple, spatial plots (x,y,z vs, nExEL, nContEH)
    • y vs nContEH : see this elog and this elog
  • Check hypothesis that the decay modifies the nContEH part of the track by making requirements that the Michel is upward going
    • lambda vs gondola | nContEH : see this elog and this elog

  • Use the event display to examine nContEH >= 8. Are these real or do they have funny topologies?
    • event display for nContEH >= 8: here
  • Resolve the difference between lambda vs nExEL | nContEH = 1 and lambda vs nCont = 1 | nExEL <=18
  • Determine how the head cut is stored
    • Answer: IsTrack() and other oneline fits, etc only stored when el_max - eh_max is small enough

Outstanding Tasks / Questions

  • Need to perform a monte carlo to understand the banks bump equivalent for the LESS mode
>
>
Please go to https://muon.npl.washington.edu/twiki/bin/view/Main/AnalysisTaskListOct22

Revision 12008-10-22 - BrendanKiburg

Line: 1 to 1
Added:
>
>
-- BrendanKiburg - 22 Oct 2008

The idea here is to list open questions, outline the proposed method for answering the questions, and prioritize the tasks.

Questions Generated from recent teleconference

Addressed during week of 10/15/2008

  • Double check the (y,z) distribution with nContEH ; also look at the y distribution where each z slice is normalized, to see if things are truly downward biased
    • Interesting results to both questions in this elog
  • I need to redo the shrubs where the z_electron is the z_tracked back to the tpc frame
    • First results for 1 file: elog
  • Response to Peter's elog estimating the lifetime deviation due to scatters: elog
  • Peter W. asks for cvs update with new shrub additions (best entrance flag, HasBestEntranceIgnoringSparks(), etc) : updated 10/15 , afternoon

Remain Open

  • Dave suggets start time scan for midplane gondolas, without nContEH cut. If OEH mustops leak into 1EH mustops, this should disappear at some point and become stable.
  • Peter Winter suggests a similar stop time scan for downward going ones
  • I suggest adding up all upward going guys (1-3, 16->14) and looking for gondola variations

  • For sara: Given a small abs(theta), look to see if the nCont distribution is different for different y-slices

  • Use the event display to examine nExEL >= 16. Are these scatters, sparks, something else?

  • Repeat the nExEL scan with a cut on the fit to the extra EL pointing/not pointing to the stop
    • This involves setting up some simple 2nd-level tracking to fit the extra pixels
    • Check auxptr and TTwoLine info in the mustop tree (for visualization on event display and observe available chi-square information)
  • Check hypothesis that the decay modifies the nContEH part of the track by making requirements that the Michel is upward going
  • Dave suggests that making plots versus theta (or gondola z), rather than just phi
  • Compare the muplus fits to the muminus
    • i peter refines the 7-parameter fit
    • make use of the par-scan macros
  • make simple, spatial plots (x,y,z vs, nExEL, nContEH)
    • run shrub level
    • add to the main mta analysis
  • Define the nExEL variable for nIslands > 1 (currently no extra EL pixel information is collected)
    • NExEL for island i = (nExEL - sum EL in all real islands)
    • Consider using IsNotAlone for cuts
  • Peter W. asks for cvs update with new shrub additions (best entrance flag, HasBestEntranceIgnoringSparks(), etc)
  • Simple 1 parameter model to show how distortion for a given gondola for n nContEH depends on the relative statistics in bin n and n-1
  • gond:dt:nContEH | angle of decay electron (upstream vs downstream)
  • gond:dt:nContEH | angle of decay elec | y_stop height

Addressed during week of 10/8/2008

  • make simple, spatial plots (x,y,z vs, nExEL, nContEH)
    • y vs nContEH : see this elog and this elog
  • Check hypothesis that the decay modifies the nContEH part of the track by making requirements that the Michel is upward going
    • lambda vs gondola | nContEH : see this elog and this elog

  • Use the event display to examine nContEH >= 8. Are these real or do they have funny topologies?
    • event display for nContEH >= 8: here
  • Resolve the difference between lambda vs nExEL | nContEH = 1 and lambda vs nCont = 1 | nExEL <=18
  • Determine how the head cut is stored
    • Answer: IsTrack() and other oneline fits, etc only stored when el_max - eh_max is small enough

Outstanding Tasks / Questions

  • Need to perform a monte carlo to understand the banks bump equivalent for the LESS mode
 
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