[Olympus] Simulation design paper

[Jan C. Bernauer]

Dear Michael,
On 18.02.13 13:37, Michael Kohl wrote:
> Dear Jan and Axel,
> thank you for posting your technical note on simulations for OLYMPUS.
> I have two immediate comments.
>
> 1) I agree that convoluting time-dependent effects and other effects 
> and imperfections into the MC will provide the correct ratio and TPE 
> effect. However, some "corrections" are still useful to be accounted 
> for in the measured data. For example, if the scattering angle in some 
> areas of the detectors, or at certain times, is not correctly 
> determined, then one would *not* like to "spoil" the MC to make it 
> equally wrong (while it will maintain the correct data/MC ratio), but 
> apply the necessary corrections to the data. So we still need to draw 
> the line of what should be "corrected" (e.g. beam offsets, as they 
> affect the scattering angle), versus what should be put into the MC. 
> Some such "corrections" applied to measured data will give a "clearer" 
> picture of the data, rather than making the MC appear diffuse, too. 
> This is also important in view of graphically representing the data, 
> rather than only quoting the final result for the ratio.
>
Ah, I think  what you mean is: The scattering angle is not the angle to 
the z axis, but to the incident beam direction. Yes, I totally agree, 
but I wouldn't call that a correction in that sense. I would call 
defining the scattering angle relative to the z-axis instead of the real 
beam a honest bug, at least if this effect is appreciable large.
So the point is: Make your analysis as correct as possible.
On the other hand: There are a lot of effects you can not correct 
event-by-event, but simulate event-by-event. For example external energy 
loss.

I first wrote: Correct everything you can do event-by-event in the data. 
But I think this is wrong. For example: The 12deg system could calculate 
a "correction weight" for each event, depending on the beam 
direction/position and produce a luminosity instead of a integrated 
cross section. But this is something I would certainly push in the 
simulation!

>
> 2) Quasi-random numbers: This is fine if the number of dimensions is 
> low. I remember that there are possible issue for 
> "highly-differential" cross sections. So it might be ok for OLYMPUS 
> but it should be looked at.
>
Great care has to be taken that you don't use the same "dimension" of 
the Quasi-Random number twice, and you have to know in advance how many 
you need for each event. This must be a constant!. But after that, I 
think you are save if you take the right series. Halton or sobol for 
example. But it is always good to cross check with a simulation with 
pseudo-random numbers.


Best,
Jan
>
> Best regards
>   Michael
>
>
>
> On Mon, 18 Feb 2013, Jan C. Bernauer wrote:
>
>> Hi all,
>>
>>  Please find attached the paper Axel and I wrote regarding the
>> simulation design.
>>
>> Axel will present the salient points at the meeting and I will give an
>> overview of the current state.
>>
>> The file is also linked in the wiki:
>> https://olympus-docu.hiskp.uni-bonn.de/dokuwiki/doku.php?id=ana-doc&#simulation 
>>
>>
>>
>> Best,
>> Jan
>> -- 
>> Dr. Jan C. Bernauer
>> DESY/Olympus
>> Notkestrasse 85, 22607 Hamburg, D
>> Tel: +49 40 8998 6402
>>
>> Massachusetts Institute of Technology
>> 77 Massachusetts Ave, Room 26-441
>> Cambridge, MA, 02139, USA
>> Phone:  (617) 253-6580
>> -------------- next part --------------
>> A non-text attachment was scrubbed...
>> Name: sim_paper.pdf
>> Type: application/pdf
>> Size: 197720 bytes
>> Desc: not available
>> Url : 
>> http://mailman.mit.edu/pipermail/olympus/attachments/20130218/d551f5b2/attachment.pdf
>> _______________________________________________
>> Olympus mailing list
>> Olympus at mit.edu
>> http://mailman.mit.edu/mailman/listinfo/olympus
>>
>
> +---------------------------------------------------------------------
> | Dr. Michael Kohl, Assistant Professor and Staff Research Scientist
> | Physics Department, Hampton University, Hampton, VA 23668
> | Jefferson Lab, C117, 12000 Jefferson Avenue, Newport News, VA 23606
> | Phone: +1-757-727-5153 (HU), +1-757-269-7343 (Jlab)
> | Fax:   +1-757-728-6910 (HU), +1-757-269-7363 (Jlab)
> | Email: kohlm at jlab.org
> | Cell:  +1-757-256-5122 (USA)
> |
> | Deutsches Elektronen-Synchrotron DESY, Bd. 66, Rm. 6,
> | Phone: +49-40-8998-6406, Cell: +49-171-101-1967
> +---------------------------------------------------------------------
>


-- 
Dr. Jan C. Bernauer
DESY/Olympus
Notkestrasse 85, 22607 Hamburg, D
Tel: +49 40 8998 6402

Massachusetts Institute of Technology
77 Massachusetts Ave, Room 26-441
Cambridge, MA, 02139, USA
Phone:  (617) 253-6580