Ticket #4923 (closed: fixed)
Errors of MuonAsymmetry algorithms cannot be fully correct
| Reported by: | Anders Markvardsen | Owned by: | Anders Markvardsen |
|---|---|---|---|
| Priority: | critical | Milestone: | Release 2.1 |
| Component: | Mantid | Keywords: | |
| Cc: | robert.whitley@… | Blocked By: | |
| Blocking: | Tester: | Alex Buts |
Description
Email has been sent to Steve since a formula he provided does not work when data are bunched.
The email is asking him to double check his error calculation formula.
Change History
comment:2 Changed 9 years ago by Anders Markvardsen
Remove last bunched bin if shorter than previous bins. re #4923
Changeset: 0774fffb49cbc64fc4c58e412b7ad65fb596ae3a
comment:3 Changed 9 years ago by Anders Markvardsen
Fixed a subtle bug when data are bunched in MuonAnalysis. re #4923
Changeset: 0ccfad6c6717889888fba4ce8445615cd808069e
comment:4 Changed 9 years ago by Anders Markvardsen
Remove last bunched bin if shorter than previous bins. re #4923
Changeset: 0774fffb49cbc64fc4c58e412b7ad65fb596ae3a
comment:5 Changed 9 years ago by Russell Taylor
Clear compiler warning. Re #4923.
Changeset: 92a262905804f2ba0ff9cf5d2bf6b71726a95c61
comment:6 Changed 9 years ago by Anders Markvardsen
- Status changed from new to accepted
The formula turned out to be correct but it is calculating the errors as square root of the counts and in that way ignoring the actual errors in the data. Steve prefers it that way.
When bunching we did this by averaging rather than summing the counts.
Bunching with this ticket has now been changed to sum counts, which at the same time means that the current error cal in MuonAsymmetry algorithms is now valid.
To check simply plot pair asymmetry data in MuonAnalysis with increasing bunching at watch the errorbars to get smaller
comment:7 Changed 9 years ago by Anders Markvardsen
- Status changed from accepted to verify
- Resolution set to fixed
comment:8 Changed 8 years ago by Alex Buts
- Status changed from verify to verifying
- Tester set to Alex Buts
comment:9 Changed 8 years ago by Alex Buts
Errors indeed become smaller. works and looks correct visually
leave exact calculations to a scientist
comment:11 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 5769
Fixed a subtle bug when data are bunched in MuonAnalysis. re #4923