Ticket #6230 (closed: duplicate)
Modify absolute normalization to produce correct cross-section
| Reported by: | Alex Buts | Owned by: | Alex Buts |
|---|---|---|---|
| Priority: | major | Milestone: | Release 2.6 |
| Component: | Framework | Keywords: | |
| Cc: | Blocked By: | ||
| Blocking: | #5567 | Tester: | Michael Reuter |
Description
The accuracy of monovanadium integral depends on sufficient statistics collected in the monovanadium scattering experiment.
Insufficient statistics causes distribution to be non-Gaussian and Bayesian statistics needs to be deployed to estimate correct averages.
The purpose of the ticket is:
When absolute normalization is used:
1) Estimate the statistics used for monovan integration and warn the user if this statistics is insufficient.
2) use best Bayesian approach in this situation
3) Write the tests to verify the vanadium cross section is calculated properly.
Change History
comment:3 Changed 8 years ago by Alex Buts
refs #6230 Modified SumSpectra algorithm
to return statistics on the sum
Changeset: 5b2fc2d835beff2b7a6ce831334f2e2054cb41ca
comment:4 Changed 8 years ago by Alex Buts
refs #6230 Method to get min-max values on axis
to estimate integration range etc and exported these functions to python
Changeset: 996924cc88f0627f85bb4e51993b1180e5413ea4
comment:5 Changed 8 years ago by Alex Buts
refs #6230 Modified sum spectra to produce weighted sum on request
Changeset: b04a8108ea15e21f97d70aa98875647fb63d5890
comment:6 Changed 8 years ago by Alex Buts
refs #6230 Modified sum spectra
to be capable of producing weighted sum
Changeset: 1986488bac1f7bde75d2bf98c06a1ae673cf5183
comment:7 Changed 8 years ago by Alex Buts
refs #6230 working Weighted averages with more tests
Changeset: edc2659c36843849679ab6dc124adb8f9d68540d
comment:8 Changed 8 years ago by Alex Buts
refs #6230 Removed deprecated Libisis summation
Changeset: 99d0399a51a3dfbcc8188c8b8fe5f114020643e9
comment:9 Changed 8 years ago by Alex Buts
refs #6230 fixing Unix warnings
Changeset: a65ed48fd10ec5498f7db8990cb631c42c5f8724
comment:10 Changed 8 years ago by Alex Buts
refs #6230 small comments
Changeset: b6f7230906a9af60b3aca23292cdff0c6a1369b1
comment:11 Changed 8 years ago by Alex Buts
refs #6230 Modified SumSpectra algorithm
to return statistics on the sum
Changeset: 5b2fc2d835beff2b7a6ce831334f2e2054cb41ca
comment:12 Changed 8 years ago by Alex Buts
refs #6230 Method to get min-max values on axis
to estimate integration range etc and exported these functions to python
Changeset: 996924cc88f0627f85bb4e51993b1180e5413ea4
comment:13 Changed 8 years ago by Alex Buts
refs #6230 Modified sum spectra to produce weighted sum on request
Changeset: b04a8108ea15e21f97d70aa98875647fb63d5890
comment:14 Changed 8 years ago by Alex Buts
refs #6230 Modified sum spectra
to be capable of producing weighted sum
Changeset: 1986488bac1f7bde75d2bf98c06a1ae673cf5183
comment:15 Changed 8 years ago by Alex Buts
refs #6230 working Weighted averages with more tests
Changeset: edc2659c36843849679ab6dc124adb8f9d68540d
comment:16 Changed 8 years ago by Alex Buts
refs #6230 fixing Unix warnings
Changeset: a65ed48fd10ec5498f7db8990cb631c42c5f8724
comment:17 Changed 8 years ago by Alex Buts
refs #6230 small comments
Changeset: b6f7230906a9af60b3aca23292cdff0c6a1369b1
comment:20 Changed 7 years ago by Alex Buts
- Status changed from accepted to verify
- Resolution set to duplicate
there was some job done on this ticket, but it all was incorporated and completed within the changes done within ticket #6667.
I have reduced vanadium runs on all four ISIS inelastic instruments and the cross-sections they produce are correct. There are possibilities for improvements in the code and in the work-flow, but they will be addressed by separate tickets.
The results of the reduction were discussed with instrument scientists and are accepted. There is nothing to test here.
comment:21 Changed 7 years ago by Michael Reuter
- Status changed from verify to verifying
- Tester set to Michael Reuter
comment:22 Changed 7 years ago by Michael Reuter
- Status changed from verifying to closed
Yes, it looks like the changes have been in use for quite some time now.
comment:24 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 7076
