diff options
author | tlatorre <tlatorre@uchicago.edu> | 2020-04-13 11:15:17 -0500 |
---|---|---|
committer | tlatorre <tlatorre@uchicago.edu> | 2020-04-13 11:15:17 -0500 |
commit | 9ebdb5f67d092a523b05a8f4654eb40bd4627601 (patch) | |
tree | 6fb94c4ed53032493658a42411f8bcb6267277db /src/find_peaks.h | |
parent | 3550aafb5784197cd2b0a3e5c1af645175084234 (diff) | |
download | sddm-9ebdb5f67d092a523b05a8f4654eb40bd4627601.tar.gz sddm-9ebdb5f67d092a523b05a8f4654eb40bd4627601.tar.bz2 sddm-9ebdb5f67d092a523b05a8f4654eb40bd4627601.zip |
add probability of a miscalibrated channel to the likelihood
This commit adds the probability that a channel is miscalibrated and/or doesn't
make it into the event to the likelihood. This was added because I noticed when
looking at the likelihood for one very high energy event that there was a
single PMT that should have been hit that wasn't in the event and which was not
marked as bad in DQXX.
I did some testing and the addition of this term does not seem to significantly
affect that atmospheric MC or the psi values for flashers. One unexpected
improvement is that it seems that external muons are more likely to correctly
reconstruct at the PSUP with this change. I haven't determined the exact cause
but I suspect it's because there is some mismodelling of the likelihood for
muons near the edge of the detector when they exit and that adding this term
allows the likelihood to ignore these PMT hits.
Diffstat (limited to 'src/find_peaks.h')
0 files changed, 0 insertions, 0 deletions