How fussy should we be when spotting transits?

• by TheEnds

  I'm not sure whether I'm supposed to approach this with the attitude of 'only tag transits if they are a perfect, obvious example' or 'tag absolutely everything that could possibly be a transit', or anything in between.

  http://talk.planethunters.org/#/subjects/APH0000dqa?quarter=1-3

  http://talk.planethunters.org/#/subjects/APH0000fg9?quarter=1-3

  Having the most trouble with curves like these. Any tips for spotting transits in them?

  Posted September 17, 2015 4:18 AM

• by davidbundy77

  In my opinion, it is better to mark too many transits than too few. It is relatively easy for analysts to filter out false positives later, but small transits could go unnoticed if nobody tags them. Also remember that the data has already been publicly available for a few months, so all the obvious transits will already have been spotted by computer algorithms or other manual searches. Your example is a typical case which is difficult for a non-expert to interpret. I would probably mark several transits but I am nevertheless doubtful that any of these dips are transits.

  Posted September 17, 2015 7:19 PM

• by TheEnds in response to davidbundy77's comment.

  Great answer, thank you! I'll loosen up a bit and mark more stuff

  Posted September 18, 2015 12:19 AM