Andrea-Havron / clustTMB

spatial finite mixture model R package
https://andrea-havron.github.io/clustTMB/
4 stars 0 forks source link

clustTMB

R-CMD-check codecov

clustTMB fits finite mixture models that incorporate both covariate and random effect structure within either the gating, the expert, or both components of the model. The clustTMB model is built in Template Model Builder (TMB) and implemented in R using formula syntax similar to other finite mixture modeling packages such as mclust. The TMB structure of the finite mixture model allows for fast computations of random effects by using the Laplace Approximation to calculate the marginal likelihood of fixed effects after integrating out random components. Spatial random effects are estimated through approximations of the Gaussian Markov Random Fields via the FEM-SPDE method introduced by INLA.

Installation

You can install the development version from GitHub with:

remotes::install_github("Andrea-Havron/clustTMB", build_vignettes = TRUE)

Example

A basic example is illustrated with the faithful dataset:

library(clustTMB)
data("faithful")
m1 <- clustTMB(response = faithful, covariance.structure = "VVV")

Vignettes

For more detailed examples, see vignettes:

## Not Run

# install.packages("utils")
utils::browseVignettes("clustTMB")

## End Not Run