spaMM

Mixed-Effect Models, Particularly Spatial Models

Inference based on mixed-effect models, including generalized linear mixed models with spatial correlations and models with non-Gaussian random effects (e.g., Beta). Both classical geostatistical models, and Markov random field models on irregular grids, can be fitted. Variation in residual variance (heteroscedasticity) can itself be represented by a generalized linear mixed model. Various approximations of likelihood or restricted likelihood are implemented, in particular h-likelihood (Lee and Nelder 2001 <doi:10.1093/biomet/88.4.987>) and Laplace approximation.

Tests Vignettes

Available Snapshots

This version of spaMM can be found in the following snapshots:

Dependencies

Imports/Depends/LinkingTo/Enhances (13)
  • proxy
  • Rcpp >= 0.12.10
  • nloptr
  • minqa
  • pbapply
  • crayon
  • gmp >= 0.6.0
  • ROI
  • Rcpp
  • RcppEigen >= 0.3.3.5.0
  • R
  • multcomp
  • RLRsim
  • Suggests (16)
  • maps
  • testthat
  • lme4
  • rsae
  • rcdd
  • pedigreemm
  • foreach
  • future
  • future.apply
  • multilevel
  • Infusion
  • IsoriX
  • blackbox
  • RSpectra
  • ROI.plugin.glpk
  • memoise >= 2.0.0
  • Version History