[Hps-analysis] MOUSE cuts

Sun Dec 9 21:54:49 EST 2018

Thanks Miriam,

> It depends how many FEEs pass the other FinalStateParticles cuts -- 
> and when the other cuts are set to their default values, this number 
> is very low, so the statement is basically false.

Other two cuts in the FinalStateParticle, are "GoodnessOfPID < 10" and 
|t_trk - t_cl - offset| < 6ns,
both of these are quite loose cuts, and I think they should not cut most 
of FEEs;
Any way these other two cuts are just particle quality cuts, if particle 
fails "GoodnessOfPID" or "Delta_tr trk_clust", then
most probably they will not be used for FEEs too.

To try to demonstrate it, I run over pass1 dsts where 
finalstateparticles have all particles in them.

In the plot below by blue shown electrons that pass Both the 
|GoodnessOfPID < 10.| and |t_trk - t_cl - offset| < 6ns.
and by red shown electrons that failed one of above cuts.

This show that around 70%+ of FEEs will go into "FinalStatParticles" 
collection.

Rafo

Rafo

On 12/9/18 9:15 PM, Diamond, Miriam wrote:
>
> Hi,
>
>
> It is correct that if an electron passes all the "FinalStateParticles" 
> cuts, then there is no way for it to be in the "OtherElectrons".
>
>
> If you put the electron momentum cut very high (e.g. at 2.1):  it is 
> still very possible that a 2.0 GeV electron will fail the *other* 
> FinalStateParticles cuts. Then it will *not* go into the 
> FinalStateParticles collection, and it *will* still appear in 
> OtherElectrons.  So, the statement "loose cut on electron momentum in 
> the FinalStateParticles doesn't allow to have all FEEs electrons to be 
> in OtherElectrons" is not necessarily true.  It depends how many FEEs 
> pass the other FinalStateParticles cuts -- and when the other cuts are 
> set to their default values, this number is very low, so the statement 
> is basically false.
>
>
> -- Miriam
>
> ------------------------------------------------------------------------
> *From:* Rafayel Paremuzyan <rafopar at jlab.org>
> *Sent:* Sunday, December 9, 2018 6:07:08 PM
> *To:* Diamond, Miriam; Nelson, Timothy Knight
> *Cc:* hps-analysis at jlab.org
> *Subject:* Re: [Hps-analysis] MOUSE cuts
> Hi Miriam,
>
>> The other MOUSE cuts on FinalStateParticles do indeed help cut out 
>> FEEs from this collection.  In other words, the electron momentum cut 
>> is just one of the cuts that discriminates against the FEEs, so if 
>> you loosen it, you still won't have many FEEs sneaking in.
> Thanks I agree with the above statement.
>
>> all the FEEs are in the OtherElectrons collection.  There are no 
>> momentum cuts on the OtherElectrons. 
> A question about above statement:
> There is a cut on the electron momentum for the "FinalStateParticles", 
> and as I understand if electron passes
> "FinalStateParticles" cuts, then there is no way it to be in the 
> "OtherElectrons" in other word's
> there is effectively a cut on "OtherElectrons" Is that right?
>
> Now in this loose MOUSE cuts, I want to put this electron momentum cut 
> high like 2.1 or high, and I think
> this means in the "OtherElectrons" there will not be electrons below 
> 2.1 GeV, am I right?
>
> If so, then what I am thinking "OtherElectrons" will not be very 
> effective for FEE studies, since for FEEs analyzer
> would like to have the whole (3\sigma+) of Gaussian part. Therefore In 
> iss361, I put the electron momentum cut 2.6,
> that all (most of) electrons will be in the "FinalStateParticles" 
> collection.
>
>> I thought I emphasized this several times in multiple summary talks, 
>> but I'll state it again here: FinalStateParticles are for making V0s 
>> out of.  OtherElectrons are for FEEs and Mollers.
>>
> I understand this statement that "FinalStateParticles" are for v0s, 
> and "OtherElectrons" for Moeller ans FEEs, however,
> in my understanding loose cut on electron momentum in the 
> "FinalStateParticles" doesn't allow to have All FEEs electrons to be 
> in "OtherElectrons".
>
> Rafo
>
>
> On 12/9/18 8:14 PM, Diamond, Miriam wrote:
>>
>> Hi,
>>
>>
>> V0s are made out of electrons and positrons in 
>> the FinalStateParticles collection. There shouldn't be any FEEs at 
>> all in this collection -- all the FEEs are in the OtherElectrons 
>> collection.  There are no momentum cuts on the OtherElectrons.  Note, 
>> there is no overlap between the FinalStateParticles collection and 
>> the OtherElectrons collection.
>>
>>
>> The other MOUSE cuts on FinalStateParticles do indeed help cut out 
>> FEEs from this collection.  In other words, the electron momentum cut 
>> is just one of the cuts that discriminates against the FEEs, so if 
>> you loosen it, you still won't have many FEEs sneaking in.
>>
>>
>> I thought I emphasized this several times in multiple summary talks, 
>> but I'll state it again here: FinalStateParticles are for making V0s 
>> out of.  OtherElectrons are for FEEs and Mollers.
>>
>>
>> -- Miriam
>>
>> ------------------------------------------------------------------------
>> *From:* Nelson, Timothy Knight
>> *Sent:* Wednesday, December 5, 2018 10:15:40 PM
>> *To:* Rafayel Paremuzyan
>> *Cc:* hps-analysis at jlab.org <mailto:hps-analysis at jlab.org>; Diamond, 
>> Miriam
>> *Subject:* Re: [Hps-analysis] MOUSE cuts
>> Thanks Rafo,
>>
>> I think these are the right questions.  Going a bit looser is OK as 
>> long as it doesn’t blow up volume. The key questions for me were:
>>
>> 1) I had thought there were more FEE contributing so that it would up 
>> the V0 collection. The fact that this didn’t happen made me wonder 
>> whether there was a problem somewhere in Rafo’s procedure relative to 
>> how these cuts were supposed to be applied.
>> 2) If it’s really true there are no FEE collected here, then isn’t 
>> that a problem in itself?  I thought the OtherElectrons would be 
>> useful for FEE studies. Again… this doesn’t fit with what I expected 
>> based on previous presentations.
>>
>> Hopefully Miriam can illuminate this when she comes up for air.
>>
>> Cheers,
>> Tim
>>
>> > On Dec 4, 2018, at 12:43 PM, Rafayel Paremuzyan <rafopar at jlab.org> 
>> <mailto:rafopar at jlab.org> wrote:
>> >
>> > Hi Miriam,
>> >
>> > Today in the analysis meeting, I was trying to justify that we can 
>> loosen some of MOUSE cuts to be safer in the future analysis step,
>> > while these cut loosening together increase number of events about 10%.
>> >
>> > There was comments that in particular making electron momentum 
>> looser would pick up a lot of FEE events, making
>> > file size to be significantly larger, however when I apply loose 
>> MOUSE cuts on pass1 data (which doesn't have MOUSE cuts),
>> > there is tiny contribution from FEE peak.
>> >
>> > I think this happens because other MOUSE cuts already cut out most 
>> of FEE peaks (Mostly e-e+ be in opposite half and P_sum < 2.8 cuts),
>> > however you are the best person to comment on this, whether the 
>> figure shown in Slide 3 of the following document is reasonable for you?
>> > 
>> https://docs.google.com/presentation/d/1rZJZLh8H0QCzEUCnDxTxFjbMnRhoawH6rxMlTUxbzAs/edit?usp=sharing 
>> <https://urldefense.proofpoint.com/v2/url?u=https-3A__docs.google.com_presentation_d_1rZJZLh8H0QCzEUCnDxTxFjbMnRhoawH6rxMlTUxbzAs_edit-3Fusp-3Dsharing&d=DwMF-g&c=lz9TcOasaINaaC3U7FbMev2lsutwpI4--09aP8Lu18s&r=0HDJrGO9TZQTE97J9Abt2A&m=YFc2llMTWRBBMx2a3SENxP0V6jBW_KGnP3C-xca_imU&s=egcClPd12dF0PxTZUuBfk4X6Jowm-xgSUDtbWubSDeg&e=>
>> >
>> > In these slides blue histograms are from pass1, when I applied 
>> loosened MOUSE cuts, and the red one represents tpass2 which is 
>> reconstructed
>> > with Original MOUSE cuts.
>> >
>> > Other's please comment, if I didn't describe questions accurately.
>> >
>> > Rafo
>> > _______________________________________________
>> > Hps-analysis mailing list
>> > Hps-analysis at jlab.org <mailto:Hps-analysis at jlab.org>
>> > https://mailman.jlab.org/mailman/listinfo/hps-analysis
>>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/hps-analysis/attachments/20181209/17e273ec/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: neiombiocgobcjof.png
Type: image/png
Size: 20388 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/hps-analysis/attachments/20181209/17e273ec/attachment-0001.png>