Register for the ACEMS 3-day short course on INLA, presented by Dr.Daniel Simpson from the Department of Statistics, University Warwick.
The aim of the course is to give an introduction to the R package R-INLA and to introduce the participants to computationally efficient spatial statistics models.
This R package provides an easy way to do Bayesian inference for latent Gaussian models using the INLA method of Rue et al. (2009) and is freely available from www.r-inla.org. The latent Gaussian models include many well-known models such as the generalized linear models, but also complex spatial, temporal and spatio-temporal models. Applications range from longitudinal data analysis and time series analysis to disease mapping, spatial survival analysis and many more.
Two major benefits over traditional Markov chain Monte Carlo algorithms are that precise estimates are available in seconds or minutes that only simple code is required by the user. The “formula” framework of R is used to handle a wide variety of models in a familiar and streamlined way which only requires small changes in the code to add or remove random effects, temporal effects, spatial effects and so on. The package has also a set of non-standard extension that allows a wide range of models to be fit, and in total, this makes the R-INLA package very attractive for Bayesian inference.