[Sane-analysis] Some results.

[Whitney R. Armstrong]

Hello Everyone,

I wanted to show some *preliminary* results that I have recently produced
and get some feedback.

Here I have calculated x^2g1 and (3/2)x^2g2  which are the two integrands
to get d2p/2.
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/x2g1g2_47_22.png
This is just the 4.7 GeV data and I will have results on the 5.9 GeV soon.
These results include:
1.3 GeV minimum track energy
Cherenkov window cut
Background contamination
Elastic radiative tail subtraction

Here are g1 and g2 vs x:
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/g1g2_47_16.png

Here is the d2p integrand from the plots above.
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/x2g1g2_47_d2_22.png

Again, these are just the results for 4.7 GeV data.

Here are A1 and A2 vs W:
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/A1A2_47_W_noel_22.png
vs x:
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/A1A2_47_noel_22.png


For comparison, in the links below you will find similar plots, however,
before considering any background or subtracting the elastic tail.
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/compare_x2g1g2_47_22.png
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/compare_g1g2_47_22.png
http://quarks.temple.edu/~whit/SANE/analysis_main/results/asymmetries/compare_A1A2_47_W_22.png


Some things I am still working:

 - Internal Inelastic Radiative Tail
   My new code produces a cross section which is much larger than the
"equivalent radiator" result and slightly larger than the Fortran POLRAD
code. You can see the results here
http://quarks.temple.edu/~whit/SANE/analysis_main/results/cross_sections/inclusive/compare_IRT.png.
Of course there is a slight model difference between the born cross
sections between the Fortran and C++ code but I don't think this is the
cause of the differences.

- Background from lower energy events
  As Oscar has shown (
https://userweb.jlab.org/~rondon/analysis/asym/world/d2_Ef-W.pdf) there is
a significan't contribution from background at lower energies. My
background fit right now is crude at best and needs improved.

- Contamination from background
  Within the analysis above the background is assumed to have no asymmetry
and is thus a dilution. However, I have been working on calculating the
asymmetry from MAID. Here is a crude result
http://quarks.temple.edu/~whit/SANE/analysis_main/results/RC/MAID_pi0_asymmetry.png
(you can ignore the red line). The bottom shows the asymmetry while the top
are the cross sections. I need to double check the jacobian (which is
rather complicated) and to improve the integration over the electron
angles.That said, an asymmetry on the order of 5% seems to persist. This is
probably less improtant for the parallel data but could have a large impact
on the perpendicular asymmeties. Also, better understanding this allows for
the lower energy data to become useful.

 - Phi binning.
  The results above are not binned in phi, I just use the kinematic
average. I made the phi bins too small to be effective. I will make them
larger (fewer in number) and produce a comparison which will hopefully not
be too different from taking the average.


Again, I will have 5.9 GeV results available soon which have more
statistics than the 4.7GeV.

Any questions, feedback or comments are appreciated. Thanks for your
attention.

-Whit


-- 
Whitney Armstrong
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/sane-analysis/attachments/20131020/1e32b396/attachment.html