In response to our call for references to successful research using Stan, Matthieu Authier points us to this:
@article{
year={2014},
journal={Biodiversity and Conservation},
volume={23},
number={10},
doi={10.1007/s10531-014-0741-3},
title={How much are stranding records affected by variation in reporting rates? A case study of small delphinids in the Bay of Biscay},
url={http://dx.doi.org/10.1007/s10531-014-0741-3},
keywords={Monitoring; Marine mammal; Strandings},
author={Authier, Matthieu and Peltier, Hélène and Dorémus, Ghislain and Dabin, Willy and Van Canneyt, Olivier and Ridoux, Vincent},
pages={2591-2612},
}
Next stop, flying squirrels!
In next week’s Harper’s Findings:
Scientists concluded that the negative binomial distribution proved useful and interpretable for modelling small delphinid strandings.
As a distribution that has two parameters, the negative binomial may make some sense, but as a distribution where it’s the number of binomial events of type A until k events of type B occur… not so much.
The interpretation as a continuous mixture of poisson processes with rates from gamma distributions seems more likely to be the real reason this distribution shows up so frequently.
Fulltext: https://pdf.yt/d/hp_iyJCfIYyyxS2D / https://dl.dropboxusercontent.com/u/85192141/2014-authier.pdf / http://libgen.org/scimag7/10.1007/s10531-014-0741-3.pdf
Pingback: Stan comes through . . . again! - Statistical Modeling, Causal Inference, and Social Science Statistical Modeling, Causal Inference, and Social Science