Fits a parametric model to a empirical variogram and estimates covariance parameters. Aditionally all fitted variogram models are plotted for verification purpose.
fit.tracevariog(emp.trace.vari, models, sigma2.0, phi.0,
fix.nugget=FALSE, nugget=0,
fix.kappa=TRUE, kappa=0.5,
max.dist.variogram=NULL)empirical trace-variogram. An object returned from the trace.variog function.
a character vector of correlation function names used in geoR against which empirical trace variogram will be fitted.
initial value for the covariance parameter \(\sigma^2\) (partial sill).
initial value for the covariance parameter \(\phi\) (range).
logical, indicating whether the nugget parameter should be estimated or not.
value for the nugget parameter.
logical, indicating whether the kappa parameter should be estimated or not.
value of the smoothness parameter.
a numerical value defining the maximum distance considered when fitting the variogram.
A list with the following components:
choosed theoretical variogram model
vector of all fitted theoretical variogram models
Variogram models and parameters
When the cov.model parameter is NULL a function determines the optimal model between spherical, exponential gaussian and matern using the received parameters. The arguments sigma2.0 and phi.0 are used as initial values for fitting each variogram model.
The parameters fix.nugget, nugget, fix.kappa, kappa and max.dist.variogram are the same for each variogram model specified in models.
Giraldo, R. (2009) Geostatistical Analysis of Functional Data. Ph.D. thesis. Universitat Politecnica de Catalunya.
Giraldo, R., Delicado, P. and Mateu, J. (2012) geofd: An R package for function-valued geostatistical prediction. Revista Colombiana de Estadistica. 35, 385-407.