Software

spStack: Bayesian Geostatistics Using Predictive Stacking

Oct 8, 2024

Fast Bayesian inference for Gaussian and non-Gaussian geospatial models without using Markov chain Monte Carlo algorithms. This R package is written in C++ with calls to FORTRAN routines for optimized linear algebra operations. Core functions include -

• spLMexact(): spatial linear model with fixed values of process parameters
• spLMstack(): spatial linear model using Bayesian predictive stacking
• spGLMexact(): spatial GLM with fixed values of process parameters
• spGLMstack(): spatial GLM using Bayesian predictive stacking

The stable version is available on CRAN.