(animal) Move Persistence Mixed-effects Models
This is a development package that fits a random walk with time-varying move persistence (autocorrelation) to multiple individual animal tracks. Move persistence (gamma_t) is a latent variable that is estimated as a linear mixed-effects function of user-supplied covariates. The models are fitted using maximum likelihood estimation via the TMB
(Template Model Builder) package and using C++. Random effects are assumed to be Gaussian. The mpmm
package leverages tools from the lme4
and glmmTMB
packages, adopting their formula syntax for random effects components.
Currently, for testing and evaluation purposes only. You will need C++ compile tools.
On PC's running Windows, ensure you have installed Rtools
On Mac's, ensure you have installed Xcode and Xcode developer tools. If installation is needed, make sure you start Xcode after install to ensure final setup of developer tools is completed. Both Xcode and Xcode developer tools can be installed from the Mac App Store
mpmm
development version is available via:
devtools::install_github("ianjonsen/mpmm")