[Sane-analysis] Follow up on Re: Sane Meeting at 4pm Wed Nov 12th in f226

[O. A. Rondon]

Hi Whit,

Here are some links on the Kolmogorov-Smirnov test to compare data sets
and test if they are different or not. I'm sure there are other more
in-depth references and tools, but this one is quite practical
http://www.physics.csbsju.edu/stats/

I applied it last year to compare Hovhannes A1 and A2 results to SLAC A1
(excluding E155 10.5 deg) and A2. (I'm also forwarding my related email
on the  subject)
https://userweb.jlab.org/~rondon/analysis/asym/world/K-S_tests/

The conclusions were that the A2 sets were quite similar, but the A1
sets were not. I think I can now understand the conclusion for A1 in
terms of SANE's limited 1.5< W <3 GeV range, which concentrates A1 to be
between 0.35 < A1 < 0.78, while the SLAC A1 data go from ~0 to close to
1, for 1.5< W <8.5 GeV.

I have added a new test of SLAC all-10.5 (data set 1) vs 10.5 deg only
(data set 2), which shows they are different,
https://userweb.jlab.org/~rondon/analysis/asym/world/K-S_tests/kolm_a1_slac-10_5.pdf
(10.5 deg. are dashed line)
https://userweb.jlab.org/~rondon/analysis/asym/world/K-S_tests/kolm_a1_slac-10_5_fraction.pdf

thus confirming my approach of not including those data, which was based
on the clear improvement of the global fit Q2 without the 10.5 deg
data, see p. 1-3 here
https://userweb.jlab.org/~rondon/analysis/asym/world/fits_143_155_13.pdf

I think you could try using the test on your RCS vs Protvino A180 at 4.7
GeV (and maybe on the other three subsets, too) to see if it can tell us
something about the differences at low x. For example, you could also
compare each BigCal segment data for E'>1.3 vs E'<1.3, and other
possible combinations.

Cheers,

Oscar

Whitney R. Armstrong wrote:
> > Hello Everyone,
> >
> > I invite you to read my post ahead of todays meeting since it is rather
> > lengthy.
> >
> > http://quarks.temple.edu/~whit/SANE/sane_meetings/11-12-2014/
> >
> > Your comments are appreciated.
> >
> > Cheers,
> > Whit
> >







-------------- next part --------------
An embedded message was scrubbed...
From: "O. A. Rondon" <or at virginia.edu>
Subject: A2  fit error band and d2 errors
Date: Tue, 17 Dec 2013 23:27:24 -0500
Size: 3357
Url: https://mailman.jlab.org/pipermail/sane-analysis/attachments/20141117/1a1dbdaf/attachment.eml