[MARMAM] mark-recapture and machine-learning: new pre-print

Robert Rankin robertw.rankin at gmail.com
Mon May 9 18:58:42 PDT 2016


For those of you who study marked-animals, I would like to bring to your
attention a pre-print manuscript about a new mark-recapture method rooted
in machine-learning. A free pdf is available here:

The manuscript proposes a new way to fit high-dimensional mark-recapture
models, using ideas from the machine-learning community. It is an
alternative "model parsimony" strategy to AICc model-averaging or
model-selection. It is targeted at researchers who seek automatic variable
selection, interaction detection, and model parsimony based on
predictive-performance. Code and tutorial are available on Github.

Rankin RW (2016) "EM and component-wise boosting for Hidden Markov Models:
a machine-learning approach to capture-recapture". bioRxiv pre-print
doi:10.1011/052266, URL:http://github.com/faraway1nspace/HMMboost

ABSTRACT: This study presents a new boosting method for capture-recapture
models, routed in predictive-performance and machine-learning. The
regularization algorithm combines Expectation-Maximization and boosting to
yield a type of multimodel inference, including automatic variable
selection and control of model complexity. By analyzing simulations and a
real dataset, this study shows the qualitatively similar estimates between
AICc model-averaging and boosted capture-recapture for the CJS model. I
discuss a number of benefits of boosting for capture-recapture, including:
i) ability to fit non-linear patterns (regression-trees, splines); ii)
sparser, simpler models that are less prone to over-fitting, singularities
or boundary-value estimates than conventional methods; iii) an inference
paradigm that is routed in predictive-performance and free of p-values or
95% confidence intervals; and v) estimates that are slightly biased, but
are more stable over multiple realizations of the data. Finally, I discuss
some philosophical considerations to help practitioners motivate the use of
either prediction-optimal methods (AIC, boosting) or model-consistent
methods. The boosted capture-recapture framework is highly extensible and
could provide a rich, unified framework for addressing many topics in
capture-recapture, such as spatial capture-recapture, individual
heterogeneity, and non-linear effects.

Thank you :)

Rob Rankin
PhD Candidate
Cetacean Research Unit
Murdoch University
Western Australia

"You could give Aristotle a tutorial. And you could thrill him to the core
of his being ... Such is the privilege of living after Newton, Darwin,
Einstein, Planck, Watson, Crick and their colleagues."
-- Richard Dawkins
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.uvic.ca/pipermail/marmam/attachments/20160509/0e7776eb/attachment.html>

More information about the MARMAM mailing list