[Hadstruct] covariance matrix
Konstantinos Orginos
kostas at jlab.org
Tue Aug 18 12:34:47 EDT 2020
Hi,
My computer was not responding during the call so I could not show you these pictures. Since some of you asked about the covariance matrix for Colin’s real part matrix elements, here it is:
[cid:49483C71-87A3-4966-8EB8-A8D6BB5D76FB]
This is the normalized covariance (ie. normalized so that the diagonal elements are 1) which is all you need to see correlations.
As you can see it has strong correlation and a nice looking pattern. The data are for z=1..8 and p=1..6, a total of 48 points flattened in vector of length 48 with p varying fast.
I do not understand the pattern I see… it would be nice to understand it!
Also it would be interesting to see how it changes under jackknife.
Here is the covariance matrix for the 3 parameters of the pdf (alpha beta gamma … I did not use a delta.)
[cid:1D125A69-440B-4674-A560-E2E58BAB2E6A]
Again you can see the strong anti-correlation between alpha and gamma.
Cheers,
Kostas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/hadstruct/attachments/20200818/49b3dcc1/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: attachment.png
Type: image/png
Size: 45235 bytes
Desc: attachment.png
URL: <https://mailman.jlab.org/pipermail/hadstruct/attachments/20200818/49b3dcc1/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: attachment.png
Type: image/png
Size: 11724 bytes
Desc: attachment.png
URL: <https://mailman.jlab.org/pipermail/hadstruct/attachments/20200818/49b3dcc1/attachment-0003.png>
More information about the Hadstruct
mailing list