GRaF (version 0.1-12)

GRaF-package: Species distribution modelling using latent Gaussian random fields

Description

GRaF fits Gaussian random field species distribution models using Laplace approximations.

Details

Package:

GRaF

Type:

Package

Version:

0.1-12

Date:

2014-01-08

License:

GPL (>= 2)

GRaF provides functions to fit, visualise, and predict from Gaussian random field models for binary data, with an emphasis on species distribution models. The principle function is graf which returns a graf object, which can be manipulated using plot, print and predict methods among others.

GRaF is under active development and future releases will enable a range of additional methods, which may (or may not) include: models for count data, spatially and temporally explicit models, the ability to limit interactions between covariates, multi-species models and input and output of raster imagery. Please feel free to contact the maintainer with bug reports and suggestions!

The development version of this package is hosted on github (https://github.com/goldingn/GRaF) and a brief tutorial is available here: https://rawgithub.com/goldingn/intecol2013/master/tutorial/graf_workshop.html

References

Golding & Purse (2013) GRaF: Fast and Flexible Bayesian Species Distribution Modelling Using Gaussian Random Fields. Manuscript in preparation. (preprint as Chapter 5 in http://dx.doi.org/10.6084/m9.figshare.767289)

See Also

graf