|
|
|
## Attendance
|
|
|
|
|
|
|
|
Alan, Jan, Deirdre, Soumen, Arun, Sayantani, Saleem and Ish
|
|
|
|
|
|
|
|
## Agenda
|
|
|
|
|
|
|
|
- Injection analysis
|
|
|
|
- Slides: [link](https://docs.google.com/presentation/d/1QgrvsgLzkeFsoYVNovqGhCcZqB8co6pD5DWnmDTn8BM/edit?usp=sharing)
|
|
|
|
|
|
|
|
## Minutes
|
|
|
|
|
|
|
|
- Minute taker: Ish
|
|
|
|
- Minutes from last call: [link](https://git.ligo.org/sayantani.datta/pca_tgr/-/wikis/April%2014,%202023)
|
|
|
|
|
|
|
|
Sayantani- Should we do the same injections we are doing for TIGER with FTA, apart from the real event analysis?
|
|
|
|
- Alan- Since we are only doing inspiral-only analysis right now, and since TIGER and FTA will be doing the same thing (in essence), we can use similar injections for TIGER and FTA.
|
|
|
|
- Jan- FTA should be similar to TIGER but FTA cannot handle precessing spins. In the past, TIGER and FTA results have agreed with each other and usually, tests are done with the two pipelines to check for systematics.
|
|
|
|
|
|
|
|
Sayantani- Which waveform will be used for FTA in O4?
|
|
|
|
- Jan- EOB ROM waveforms. Allows for testing of systematics when changing the underlying waveform model (i.e., comparison with IMRPhenom family).
|
|
|
|
|
|
|
|
Sayantani- Is it okay if we use IMRPhenomXP for TIGER and EOB ROMs for FTA?
|
|
|
|
- Alan- For now, yes. But later, we might have to work with other waveform models to investigate systematics.
|
|
|
|
|
|
|
|
Soumen- XP is a precessing waveform model and EOB ROMs are aligned spin. So, better to use IMRPhenomXAS for comparison.
|
|
|
|
- Arun- TGR chairs might want us to use a processing waveform. So, while you run on XAS, run on XP as well.
|
|
|
|
- Sayantani- Then we can do the same for real events?
|
|
|
|
- Arun- We should use XP for real events, which is what is usually done, and wait for TIGER and FTA to resolve the precessing vs aligned-spin issues.
|
|
|
|
|
|
|
|
Alan- You are making the assertion that for GR signals you will not see a false deviation. Need to do injection runs to convince ourselves that this is indeed the case.
|
|
|
|
|
|
|
|
Alan- Visualizing the off-diagonal elements on the combined PCA and color coding will help see if the combined PCA estimates are intermediate to the ones obtained from individual events
|
|
|
|
- Saleem- Overplotting corner plots with combined and single events might serve the purpose.
|
|
|
|
|
|
|
|
|
|
|
|
## Action items
|
|
|
|
|
|
|
|
1. Add links for inspiral-SNR calculation scripts and their relevant reviews.
|
|
|
|
2. Update slides with background information about TIGER and FTA, including latest documentation, waveform used, etc. |
|
|
\ No newline at end of file |
