README.md

R package occuHMM

What it does

This is my first R package, let's see how it goes. It contains R functions to simulate data and fit several occupancy models in the hidden Markov modelling (HMM) framework. Real data will also be included.

What it does not do

This is just a repository that contains R functions to fit occupancy models. To fit these models in a real-world applications, I warmly recommend you use unmarked or, if you're happy to leave R, `PRESENCE, MARK or E-SURGE. Note that MARK can be called from R using RMark.

To install the package

install.packages("devtools")
library("devtools")
install_github('oliviergimenez/occuHMM')

Which occupancy models?

The prefix fit is for fitting a model, sim is for simulating data from a model and dev is for the function that gives (half) the deviance (for internal use, but you're free to look into it). The suffix colext is for dynamic models (which I stole from unmarked), colextfp for dynamic models with false positives.

A session example

vignette("colext_occuHMM")

Motivation and acknowledgments

My first attempt to formulate occupancy models as HMM dates back to Gimenez, O., L. Blanc, A. Besnard, R. Pradel, P. F. Doherty Jr, E. Marboutin and R. Choquet (2014). Fitting occupancy models with E-SURGE: hidden-Markov modelling of presence-absence data. Methods in Ecology and Evolution. 5: 592–597. We used program E-SURGE, which is developed by R. Choquet in our research team.

Despite its flexibility, E-SURGE is a Windows standalone executable, which impedes reproducibility (at least partly). The idea of building a R package arose from feedbacks I got on a 1-day workshop I gave twice in 2016 using E-SURGE (in Montpellier and Seattle for ISEC2016). Big thanks to the attendees!



oliviergimenez/occuHMM documentation built on May 24, 2019, 12:52 p.m.