X-Pipeline Review Meeting Monday 8 June 2009 09:00 Pacific / 12:00 Eastern
Agenda: Monday 8 June 2009 09:00 Pacific / 12:00 Eastern
- Current paper draft in CVS [ pdf ]
- Paper draft outstanding issues [ txt ]
- Analysis outstanding issues [ twiki ]
Minutes:Attendance: Jolien, Patrick, Xavi
- Binomial test issue
- drop section on binomial test; report binomial probability for anything with p_local below 5%; show that it is wholy consistent with the null hypothesis.
- In table, give sample size in addition to p_local. Typically around 500 background jobs for the top candidates. Smallest was p_local=7/537 with the extra background jobs (1.3%).
- Comments on the paper
- Do we know what this event was? Beyond putting some words in the paper, we should look at Q-scans of the top five candidates. All had null streams. Detection Checklist Notes.
- Loudest event: lies on edge of background cloud and on the edge of the injection cloud. Seems to be a 60 Hz event. Think that a q-scan would be illuminating. This is an 8 Hz time resolution event, which has frequency support between 60 Hz and 68 Hz. (Lowest central frequency is 64 Hz.) 9 TF pixels; fairly long duration. Reasonably strong cut is nullIoverE of -3. Seems to be some crappy data, which led to a strong cut. Loudest off-source events shows a shoulder that indicates some junk in the background.
- CBC found more redshifts that Isabel doesn't believe. Need to make consistent.
- Definition of polarization frame. Address the issues there.