GeoCV function

n-fold kriging Cross-validation

The procedure use the GeoKrig or GeoKrigloc function to compute n-fold kriging cross-validation using informations from a GeoFit object. The function returns some prediction scores. UTF-8

GeoCV(fit, K=100, estimation=TRUE, optimizer=NULL,
     lower=NULL, upper=NULL, n.fold=0.05,local=FALSE,
    neighb=NULL, maxdist=NULL,maxtime=NULL,sparse=FALSE,
    type_krig="Simple", which=1, parallel=FALSE, ncores=NULL)

Arguments

• fit: An object of class GeoFit.
• K: The number of iterations in cross-validation.
• estimation: Logical; if TRUE then an estimation is performed at each iteration and the estimates are used in the prediction. Otherwise the estimates in the object fit are used.
• optimizer: The type of optimization algorithm if estimation is TRUE. See GeoFit for details. If NULL then the optimization algorithm of the object fit is chosen.
• lower: An optional named list giving the values for the lower bound of the space parameter when the optimizer is L-BFGS-B or nlminb or optimize if estimation is TRUE.
• upper: An optional named list giving the values for the upper bound of the space parameter when the optimizer is L-BFGS-B or nlminb or optimize if estimation is TRUE.
• n.fold: Numeric; the percentage of data to be deleted (and predicted) in the cross-validation procedure.
• local: Logical; If local is TRUE, then local kriging is performed. The default is FALSE.
• neighb: Numeric; an optional positive integer indicating the order of neighborhood if local kriging is performed.
• maxdist: Numeric; an optional positive value indicating the distance in the spatial neighborhood if local kriging is performed.
• maxtime: Numeric; an optional positive value indicating the distance in the temporal neighborhood if local kriging is performed.
• sparse: Logical; if TRUE kriging and simulation are computed with sparse matrices algorithms using spam package. Default is FALSE. It should be used with compactly supported covariances.
• type_krig: String; the type of kriging. If Simple (the default) then simple kriging is performed. If Optim then optimal kriging is performed for some non-Gaussian RFs
• which: Numeric; In the case of bivariate cokriging it indicates which variable to predict. It can be 1 or 2
• parallel: Logical; if TRUE then the estimation step is parallelized
• ncores: Numeric; number of cores involved in parallelization.

Returns

Returns an object containing the following informations: - predicted: A list of the predicted values in the CV procedure;

• data_to_pred: A list of the data to predict in the CV procedure;

• mae: The vector of mean absolute error in the CV procedure;

• mad: The vector of median absolute error in the CV procedure;

• brie: The vector of brie score in the CV procedure;

• rmse: The vector of root mean squared error in the CV procedure;

• lscore: The vector of log-score in the CV procedure;

• crps: The vector of continuous ranked probability score in the CV procedure;

See Also

GeoKrig.

Author(s)

Moreno Bevilacqua, moreno.bevilacqua89@gmail.com ,https://sites.google.com/view/moreno-bevilacqua/home, Víctor Morales Oñate, victor.morales@uv.cl , https://sites.google.com/site/moralesonatevictor/, Christian", Caamaño-Carrillo, chcaaman@ubiobio.cl ,https://www.researchgate.net/profile/Christian-Caamano

Examples

library(GeoModels)

################################################################
########### Examples of spatial kriging ############
################################################################

model="Gaussian"
set.seed(79)
x = runif(400, 0, 1)
y = runif(400, 0, 1)
coords=cbind(x,y)
# Set the exponential cov parameters:
corrmodel = "GenWend"
mean=0; sill=5; nugget=0
scale=0.2;smooth=0;power2=4

param=list(mean=mean,sill=sill,nugget=nugget,scale=scale,smooth=smooth,power2=power2)

# Simulation of the spatial Gaussian random field:
data = GeoSim(coordx=coords, corrmodel=corrmodel,
              param=param)$data

## estimation with pairwise likelihood
fixed=list(nugget=nugget,smooth=0,power2=power2)
start=list(mean=0,scale=scale,sill=1)
I=Inf
lower=list(mean=-I,scale=0,sill=0)
upper=list(mean= I,scale=I,sill=I)
# Maximum pairwise likelihood fitting :
fit = GeoFit(data, coordx=coords, corrmodel=corrmodel,model=model,
                    likelihood='Marginal', type='Pairwise',neighb=3,
                    optimizer="nlminb", lower=lower,upper=upper,
                    start=start,fixed=fixed)

#a=GeoCV(fit,K=100,estimation=TRUE,parallel=TRUE) 
#a$rmse