Package index • spStack

Stacking

Implements Bayesian predictive stacking for spatial linear and generalized linear models.

spLMstack(): Bayesian spatial linear model using predictive stacking

spGLMstack(): Bayesian spatial generalized linear model using predictive stacking

Exact samplers and finds LOO-PD

Samples from the posterior distribution conditional on fixed values of spatial process parameters and finds leave-one-out predictive densities

spLMexact(): Univariate Bayesian spatial linear model

spGLMexact(): Univariate Bayesian spatial generalized linear model

Analysis of output

Stacked posterior

Helps analyze output of stacking algorithms

stackedSampler(): Sample from the stacked posterior distribution

get_stacking_weights(): Optimal stacking weights

Surface plots

Creates and plots interpolated spatial surfaces

surfaceplot(): Make a surface plot

surfaceplot2(): Make two side-by-side surface plots

Matrix algorithms

Provides compiled code for various Cholesky factor updates helpful for those designing their own cross-validation algorithms

cholUpdateRankOne() cholUpdateDel() cholUpdateDelBlock(): Different Cholesky factor updates

Synthetic datasets

Various simulated spatial data sampled on unit square

simBinary: Synthetic point-referenced binary data

simBinom: Synthetic point-referenced binomial count data

simGaussian: Synthetic point-referenced Gaussian data

simPoisson: Synthetic point-referenced Poisson count data

sim_spData(): Simulate spatial data on unit square

Other utilities

spStack spStack-package: spStack: Bayesian Geostatistics Using Predictive Stacking

iDist(): Calculate distance matrix