[G8b_run] mom corr function and histos

[Michael Dugger]

g8b run group,

Chuck: Nice job :)

The run group should come to an agreement as to the energy and momentum 
corrections. It would be best if we all used the same corrections and had 
some sort of documentation to point to.

My feeling is that a kinematic fitting routine should give the best 
results for this sort of study. Since replacing the confidence level cut 
with a missing mass cut does not change the results in any real meaningful 
way, the selection of events due to confidence level does not appear to 
be very important in the end analysis. As long as the pull distributions 
look good, I feel comfortable going with Chuck's results.

Assuming that I read the plots correctly, Chuck's results suggest that the 
photon energies in the range of interest (0.9 to 2.0 GeV) are within 0.3 
MeV ! If this is the case, we do not need to worry about the photon energy 
correction and we state the width of an eid as being the uncertainty in 
the photon energy.

Question: Should meet next week to discuss this?

Take care,
Michael

On Thu, 27 Aug 2009, Charles Hanretty wrote:

> Hello all,
>     Here is my momentum correction function and array again.  You can find
> it at:
> /u/home/hanretty/rootbeer2.1/NTuple_code/corrected_proton_mom.cc
>
> the array needed can be found at:
> /u/home/hanretty/rootbeer2.1/NTuple_code/mom_corr_array.pro
>
> The function requires you to give it the proton's 4-vec as it
> comes from the data file, no corrections should be made to this 4-vec
> prior to the call to the function.  This 4-vec is used to find the correct
> bin in momentum, theta, and phi and the appropiate correction factor.  The
> function also needs a 3-vec vertex position and the run number.  The 3-vec
> and run number are needed to apply the ELoss package to the 4-vector both
> before the corrections are applied and after. This function does not
> apply ELoss twice to the same 4-vector though.  ELoss is first applied to
> the 4-vec so the binning can be determined.  The correction factors are
> then applied to the (raw) 4-vec components and the momCorred 4-vec then
> has ELoss applied to it.  The function then returns the corrected
> 4-vec.
>
> I have also made some plots for tht tagger sag using a different
> confidence level cut and also the suggested mm^2 cut.  The first of the
> two plots I'd like to share show the taggerSag distributions using a conf
> lvl cut of 10% and 0.1%. In the histo, the lower distribution comes from
> using a conf lvl cut of 10% while the upper comes from using a conf lvl
> cut of 0.1%.  The low points of the two distributions are almost the same
> but the "humps" of the distribution using the lower conf lvl cut are more
> pronounced:
> http://hadron.physics.fsu.edu/~hanretty/Files/TaggerSag/EtrueMinusEmeas_v_Emeas_FitMeans_BothConfCuts.gif
>
> The second plot again has two taggerSag distributions.  The bottom one is
> the original dist (the one using a conf lvl cut of 10%) and the top one is
> the distribution made using the mm^2 cuts that were suggested by Mike.
> The effect of this mm^2 cut is almost exactly the same as the effect from
> making a conf lvl cut of 0.1%:
> http://hadron.physics.fsu.edu/~hanretty/Files/TaggerSag/EtrueMinusEmeas_v_Emeas_ConfCutAndMM2Cut.gif
>
> -Chuck
> _______________________________________________
> G8b_run mailing list
> G8b_run at jlab.org
> https://mailman.jlab.org/mailman/listinfo/g8b_run
>