Matthieu Authier writes:
Here is a simulation study using regularized regression with post-stratification to estimate dolphin bycatch from non-representative samples. The Stan code is accessible here.
We’ve used also RRP on a case study with samples from FR where we know that independent observers are preferentially allowed on boat when dolphin bycatch is low (a report is being written at the moment on that, and it will be the second part of the dead dolphin duology for RRP). RRP is giving more plausible estimates in this case.
For those not familiar with the jargon, “bycatch” is “the non-intentional capture or killing of non-target species in commercial or recreational fisheries.”
It’s great to see MRP and Stan being used for all sorts of real problems.
P.S. Authier has an update:
The article has been published and vastly improved thank to peer review. The simulation study is here and the case study is here.
Stan was instrumental in both cases to be able to fit the models.
The model we developed is now used to update estimate bycatch in the Bay of Biscay.
How would one summarise what this figure shows?
I’d like to hear as well. Looks like it’s a seasonal pattern of fishing effort and bycatch fraction?
The figure shows correlation between by-catch risk (estimated at the week-level) and a fishing operation duration. The large jump is partly an artefact linked to using calendar year and not using smoothing between week 52 of year t and week 1 of year (t+1). This makes little difference in pratice (we tried an analysis in which the fishing year starts in november) as very few onboard observers have been operating in December so there are very few data there.
By-catch risk is very high in week one but almost zero in week 52? Is this a model artifact? If the season was shorter than the year, an high by-catch might make sense at the opening of the season, but with continuous fishing it strikes me as odd. The same for the duration. Why does it jump from one week to the next at the end of the model year?