| RMmodelsMultivariate {RandomFields} | R Documentation |
Here, multivariate and vector-valued covariance models are presented.
Covariance models
RMbicauchy | a bivariate Cauchy model |
RMbiwm | full bivariate Whittle-Matern model (stationary and isotropic) |
RMbigneiting | bivariate Gneiting model (stationary and isotropic) |
RMbistable | a bivariate stable model |
RMcurlfree | curlfree (spatial) vector-valued field (stationary and anisotropic) |
RMdelay | bivariate delay effect model (stationary) |
RMdivfree | divergence free (spatial) vector valued field, (stationary and anisotropic) |
RMexponential | functional returning exp(C) |
RMkolmogorov | Kolmogorov's model of turbulence |
RMmatrix | linear model of corregionalisation |
RMmqam | multivariate quasi-arithmetic mean (stationary) |
RMparswm | multivariate Whittle-Matern model (stationary and isotropic) |
RMschur | element-wise product with a positive definite matrix |
RMtbm | turning bands operator |
RMvector | vector-valued field (combining RMcurlfree and RMdivfree)
|
Trend models
RMtrend | for explicite trend modelling |
R.models | for implicite trend modelling |
R.c | binding univariate trend models into a vector |
Martin Schlather, schlather@math.uni-mannheim.de
Chiles, J.-P. and Delfiner, P. (1999) Geostatistics. Modeling Spatial Uncertainty. New York: Wiley.
Schlather, M. (2011) Construction of covariance functions and unconditional simulation of random fields. In Porcu, E., Montero, J.M. and Schlather, M., Space-Time Processes and Challenges Related to Environmental Problems. New York: Springer.
Schlather, M., Malinowski, A., Menck, P.J., Oesting, M. and Strokorb, K. (2015) Analysis, simulation and prediction of multivariate random fields with package RandomFields. Journal of Statistical Software, 63 (8), 1-25, url = ‘http://www.jstatsoft.org/v63/i08/’
Wackernagel, H. (2003) Multivariate Geostatistics. Berlin: Springer, 3rd edition.
RFformula, RMmodels,
RM,
RMmodelsAdvanced
‘multivariate’, a vignette for multivariate geostatistics
RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set ## RFoptions(seed=NA) to make them all random again n <- 100 x <- runif(n=n, min=1, max=50) y <- runif(n=n, min=1, max=50) rho <- matrix(nc=2, c(1, -0.8, -0.8, 1)) model <- RMparswmX(nudiag=c(0.5, 0.5), rho=rho) ## generation of artifical data data <- RFsimulate(model = model, x=x, y=y, grid=FALSE) ## introducing some NAs ... data@data$variable1[1:10] <- NA data@data$variable2[90:100] <- NA plot(data) ## co-kriging x <- y <- seq(0, 50, 1) k <- RFinterpolate(model, x=x, y=y, data= data) plot(k, data) ## conditional simulation z <- RFsimulate(model, x=x, y=y, data= data) plot(z, data)