_ROOT Version: root_v5.34.36
I’m fitting pion yield w.r.t transverse momentum, but the covariance matrix that I got is identically 0 matrix (7-by-7, for 7 parameters). Then I added the fit option “V”, verbose mode. It tells me call reaches limit:
And right below that, we can see the EXTERNAL ERROR MATRIX is the covariance matrix I wanted, but when I print it again, it still gives me 0-matrix:
So under such circumstance, would it be promising by just increase the limit of the call (how?)? By the way, some other datasets works just fine.
I answered in your previous post:
Thanks. Good news: after using the verbose mode “V”, it returns to me some EXTERNAL ERROR MATRIX, which seems to be the covariance matrix that I always want (see the terminal report below).
But still the MIGRAD thing didn’t converge, I’m not sure how that happened when everything “seems” to be fine (the fit result seems to agree with my data; and it actually get some reasonable covariance matrix, and just doesn’t want to share it with me, when I print it.).
Let me know if this works.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.