search

pbs-assess / dogfish-assess

0 stars 0 forks source link

Revisit IPHC hook competition #6

Closed seananderson closed 5 months ago

seananderson commented 1 year ago
  1. Probably use Joe Watson's censored Poisson approach (for HBLL too). It's working well in sdmTMB now.
  2. There may be issues with the current baited hook proportions. Notes from @jdunic:

the gfiphc data accounts for chum only observed counts, but the raw data doesn't seem to include this information. And so in 2012, I think that is what explains the difference observed hook counts of about 100 - 200 hooks. I also see some big differences at 47 stations in 2006, where the gfiphc hook counts seem to have about 100 more hooks observed than the raw data. Other differences are <10. Except for a weird difference of 700 in 2018. These differences occur at stations that have a Y Eff code so I don't think they were filtered out in dogfish?

Consider just using what @jdunic creates for the synopsis report.