lgm: Linear Geostatistical Models

Description

Fits a linear geostatistical model using Maximum Likelihood Estimation

Usage

lgm(data,  cells, covariates=NULL, formula=NULL,
maternRoughness=1,fixMaternRoughness=TRUE, 
aniso=FALSE, boxcox=1, fixBoxcox=TRUE,
 nugget = 0, fixNugget = FALSE,...)

Arguments

data

An object of class SpatialPointsDataFrame containing the data.

cells

Either an integer giving the number of cells in the x direction, or a raster object which will be used for the spatial random effect. If the cells in the raster are not square, the resolution in the y direction will be adjusted to make it so.

covariates

Either a single raster, a list of rasters or a raster stack containing covariate values used when making spatial predictions. Names of the raster layers or list elements correspond to names in the formula. If a covariate is missing from the data object

formula

Fixed effects formula, defaults to a linear combination of each of the layers in the covariates object. The response variable defaults to the first variable in the data object, and formula can be an integer o

maternRoughness

Order of the Matern correlation

fixMaternRoughness

Set to FALSE to estimate the Matern order

aniso

Set to TRUE to use geometric anisotropy.

boxcox

Box-Cox transformation parameter, set to 1 for no transformation.

fixBoxcox

Set to FALSE to estimate the Box-Cox parameter.

nugget

Value for the nugget effect (observation error

fixNugget

Set to FALSE to estimate the nugget effect parameter.

...

Additional options passed to likfit

Value

If the prediction locations are identical for the fixed effects and random effects, a single raster stack is returned with the following layers:
fixedEstimated means from the fixed effects portion of the model
randomPredicted random effect
krige.varConditional variance of predicted random effect (on the transformed scale if applicable)
predictPrediction of the response, sum of fixed and random effects. If exp.pred is TRUE, gives predictions on the exponentated scale, and half of krige.var is added prior to exponentiating
predict.logIf exp.pred=TRUE, the prediction of the logged process.
If the prediction locations are different for fixed and random effects (typically coarser for the random effects), a list with two raster stacks is returned.
predict.boxcoxIf a box cox transformation was used, the prediction of the process on the transformed scale.
If the prediction locations are different for fixed and random effects (typically coarser for the random effects), a list with two raster stacks is returned.
predictionA raster stack as above, though the random effect prediction is resampled to the same locations as the fixed effects.
randomthe predictions and conditional variance of the random effects, on the same raster as cells
Additionally, returned are
parametersA data frame with parameter estimates, and standard errors of the fixed effects parameters
likfitresults from likfit

Details

Calls likfit and krige

Examples

Run this code

data(swissRain)
 
swissRes =  lgm(swissRain, cells=30, formula="rain",
	covariates=swissAltitude, boxcox=0.5, fixBoxcox=TRUE, 
	aniso=TRUE, 
	# starting values, arguments as in geoR::likfit
	ini.cov.pars=c(70, 20000), psiA=0.65, psiR=8
	)

swissRes$parameters

plot(swissRes$predict[["predict"]]) 
load(url("http://www.filefactory.com/file/frd1mhownd9/n/CHE_adm0_RData"))
plot(spTransform(gadm, swissRain@proj4string), add=TRUE)