Hi,
Before tomorrow. I took a look at the error in Abbott's fit. At 1.95fm-1,
which is right between the two points at low Q2 that I normalize to, the
fit is the following:
GC 0.1808+-0.00096
GQ 0.2066+-0.01123
GM 0.2188+-0.002789
A 0.03649
B 0.001994
all GC, GM, GQ are normalized to 1 at Q2=0. the static value must be
multiplied back before applying to T20 calculation.
the dominant error is in GQ which is a good 5%.
propagate these into T20 at that Q value, I get about +-6%. This might be
a better estimation on this piece of systematic error.
This is just a quick look on scratch paper, I will be checking my
calculations, just think it would be better to share this info with
everyone going to tomorrows meetings.
The data was composed in 2001, after than VEPP published a few points at
the dip. Looking at all the published data on T20, VEPP and possibly
BLAST, are the only two normalize to theory/world data (not sure of the
weird data point from Bonn). NIKHEF uses ion-pol, BATES and JLAB use
recoil polarimeters.
Interesting thing is that in 1990, VEPP cited +-5% systematic error in
normalization, based on Paris potential and world data of A and B
available then. In 1996, NIKHEF data fall out side of any theory curves at
that moment, and since, some theory models moved toward NIKHEF data and
widened the band. I wonder if VEPP data need some more massage in light
of this developement.
Chi
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:32 EST