[Halld-physics] [EXTERNAL] Re: Vector-Pseudoscalar WG meeting tomorrow 11/6 @ 12 pm

[Shepherd, Matthew]

During this meeting today, there was some discussion about trivial ambiguities in amplitude analysis.  I was noting that there is a function in the AmpTools framework:    FitResults::rotateResults().   The goal of this function is to establish a convention that breaks these ambiguities.  It is a non-const function so it actually modifies the production parameters and also the covariance matrix by making a series of sign flips.  WARNING:  this function seems clearly written for cartesian representation of production parameters.

The function was written by Jake Bennett and described in section 5.3.1 of his thesis here:

https://inspirehep.net/files/c0f42806cf8dc1afe267908d6464e66d

(INSPIRE page:  https://inspirehep.net/literature/1615235)

I think all aspects of that argument and the formulation translate to GlueX as well.  In BESIII fits there were multiple coherent sums due to the unknown polarization of the initial state J/psi and final state photon.

Matt

PS:  I think we were talking about Gaussian assumptions also a few meetings ago...  there are some interesting contour plots in Fig 6.17 and 6.18 that show non-Gaussian behavior away from the minimum.  When Jake did this work, we weren't aware of the idea of bootstrap and how it might be useful for uncertainty analysis.



> On Nov 5, 2023, at 9:59 PM, Justin Stevens via Halld-physics <halld-physics at jlab.org> wrote:
> 
> Hi All, 
> Our next Vector-Pseudoscalar WG meeting is scheduled for tomorrow, November 6th at 12 pm with F326 reserved at JLab.  Please find the agenda here https://halldweb.jlab.org/wiki-private/index.php/Vector_Pseudoscalar_PWA_Meeting_2023-11-6 and add to it as needed.
> 
> -Justin
> 
> 
> 
> _______________________________________________
> Halld-physics mailing list
> Halld-physics at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-physics


-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4075 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/halld-physics/attachments/20231106/40e44e4b/attachment.p7s>