[G8b_run] New Polarization Tables

[Michael Dugger]

Volker,

Your plots seem to suggest that the Delta(I_s) is well approximated by a 
Gaussian centered at zero with a standard deviation of approximately 0.03
(looks like 3 sigma is at about +/- 0.1).

If I understand correctly: 
If the I_s values were ~1.0 then the standard deviation of the 
difference between the old and new polarization would be about 0.03. 
However, if the average I_s values were about 0.5, then the standard 
deviation between the old and new values would be about 0.06. To get a 
feel for the differences in the polarizations, it might be nice to have a 
rough idea of how large the average I_s values are.

If I remember correctly, the regions where I see systematic deviations of 
between the old and new polarizations, will be in the overlaps between 
neighboring coherent edge settings. It sounds like the overlaps are the 
next thing you are looking into.

Looks like you are making good progress on the polarization study.

Thanks for keeping us informed.

Take care,
Michael



On Wed, 12 Oct 2011, Volker Crede wrote:

> Mike,
>
> you are (almost) right, plotted is Delta(I_s) = I_s_old - I_s_new (the other way around), where I_s_old is I_s for the old polarization values and I_s_new is I_s for the new polarization values. At this point, it does not matter much what the observable I_s is (it is one of the beam asymmetries for the two pions). We are just comparing "old" and "new".
>
> And yes, on average it is fair to say what you mentioned in your email.
>
>        Volker
>
>
> On Oct 12, 2011, at 2:33 PM, Michael Dugger wrote:
>
>>
>> Volker,
>>
>> I want to make sure I understand the plots shown.
>>
>> Is it fair to say that if:
>> I_s_old is the I_s for the old polarization table
>>
>> and
>>
>> I_s_new is the I_s for the new polarization table
>>
>> then what is shown in your plots is DeltaI_s, where
>> DeltaI_s = I_s_new - I_s_old?
>>
>> I am not very familiar with the I_s observable. Would it be fair to say that
>> I_s_new = I_s_old*P_old/P_new , where
>> P_old is the polarization from the old tables and
>> P_new is the polarization from the new tables?
>>
>> Thanks for your time.
>>
>> -Michael
>>
>>
>>
>> On Wed, 12 Oct 2011, Volker Crede wrote:
>>
>>> Hi all,
>>>
>>> I have a few pictures to show for the two-pion analysis comparing the old polarization tables with the new ones:
>>>
>>>    http://hadron.physics.fsu.edu/~crede/FILES/polarization.root
>>>
>>> The nomenclature of the histograms in the ROOT file is Difference_00_[energy bin] comparing the observable I_s using the new and the old values for the degree of polarization. I have used an average polarization for each energy bin. At this point, the energy bins are 50-MeV wide:
>>>
>>> 1100 - 1150 MeV        Bin 1     1.3 GeV data set
>>> 1150 - 1200 MeV        Bin 2
>>> 1200 - 1250 MeV        Bin 3
>>> 1250 - 1300 MeV        Bin 4
>>> 1300 - 1350 MeV        Bin 5     1.5 GeV data set
>>> etc.
>>>
>>> There are four different topologies (either missing a particle or detecting them all) and each energy bin is further broken up into 20*20 = 400 bins for the two angles in the rest frame of the pions: 1600 bins maximum (acceptance kills many, though) that can be compared per energy bin.
>>>
>>> I do not observe any effects resulting in a systematic shift of these distributions. All difference distributions give good Gaussian distributions; they are definitely all symmetric. The same is true (no systematic problems) when I look at individual topologies.
>>>
>>> This is all a good sign. I will further look into what Mike did for the single pion.
>>>
>>> Best wishes
>>>
>>>      Volker
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> G8b_run mailing list
>>> G8b_run at jlab.org
>>> https://mailman.jlab.org/mailman/listinfo/g8b_run
>>>
>
>
>