The goal ofbayesmove is to analyze animal movementusing a non-parametric Bayesian framework, which addresses a number oflimitations of existing segmentation methods and state-space models.This package currently offers two different model frameworks on which tomake behavioral inference from animal telemetry data: 1)themixed-membership method for movement (M4) that providessegment-level behavioral state estimation, and 2)themixture model for movement (M3) that providesobservation-level behavioral state estimation.
The M4 model is a two-stage framework that first partitionsindividual tracks into segments (via reversible-jump Markov chain MonteCarlo) and subsequently clusters these segments into latent behavioralstates (via non-parametric Latent Dirichlet Allocation). This frameworkallows the analysis of multiple telemetry and biologging data streams,which must first be discretized into a set of bins before they can beanalyzed. The observation-level M3 model also requires that data streamsare first discretized, but then directly clusters these observationstogether into behavioral states within a single step (via anon-parametric mixture model). While the outcome is similar to that fromstate-space and hidden Markov models, this observation-level model doesnot assume an underlying Markov property or use a mechanistic process(e.g., correlated random walk). Additional details about the M4 methodcan be found in Cullen et al. (2022)doi:10.1111/2041-210X.13745,while further details about the M3 method can be found in Valle etal. (2022)doi:10.1002/eap.2524.
This package also includes features to check model convergence basedon the log-likelihood for each MCMC iteration. Model output are oftenreturned in a format that istidyverse-friendly, whichallows for easy visualization usingggplot2. Additionally,a Shiny app can be launched to dynamically and interactively exploreanimal movement data, including options to filter and export data fromthe app.
You can install the latest CRAN release with:
install.packages("bayesmove",dependencies =c("Imports","LinkingTo","Suggests"))which will ensure that all packages needed to run the Shiny app areinstalled.
You can install the lateststable version of thepackage from GitHub with:
# install.packages("remotes")remotes::install_github("joshcullen/bayesmove")or latest development (unstable) version with:
# install.packages("remotes")remotes::install_github("joshcullen/bayesmove@dev")If installing from GitHub, ensure that you have a tool installed forcompiling C++ code:
For PC’s running Windows, installRtools if youhave not already done so.
For Mac’s, install theCommand Line Toolsfor Xcode by executingxcode-select --install in theterminal; or you can download the latest version from the URL (freedeveloper registration may be required). A full Xcode install uses up alot of disk space and is not required.
If you are receiving errors from the model output that you believe tobe bugs, please report them as issues in the GitHub repo. Additionally,if there are any other features you would like added to this package,please submit them to the issue tracker.