Cokriging of a primary variable from co-located primary and secondary data

Exponential direct covariances (sill minus nugget, plus the nugget at distance zero) and an exponential cross-covariance, written as a sum of coregionalization structures and solved by LmcCokriging: simple cokriging with known means (default zero means, the former behaviour) or ordinary cokriging with means = NULL. Python parity: morie.fn.cokrg.cokriging.

Usage

cokrg(
  x,
  y,
  coords,
  target,
  sill_p = 1,
  range_p = 1,
  sill_s = 1,
  range_s = 1,
  cross_sill = 0.5,
  cross_range = 1,
  nugget = 0,
  means = c(0, 0)
)

morie_cokrg(
  x,
  y,
  coords,
  target,
  sill_p = 1,
  range_p = 1,
  sill_s = 1,
  range_s = 1,
  cross_sill = 0.5,
  cross_range = 1,
  nugget = 0,
  means = c(0, 0)
)

Arguments

x: Primary variable (n,).
y: Secondary variable (n,).
coords: Co-located coord matrix.
target: Target coords (m by d) or (d,).
sill_p, range_p: Primary auto-covariance parameters.
sill_s, range_s: Secondary auto-covariance parameters.
cross_sill, cross_range: Cross-covariance parameters.
nugget: Nugget.
means: Known means c(m1, m2) for simple cokriging; NULL for ordinary cokriging.

Value

Named list: estimate, se, n, method.

References

Wackernagel, H. (2003). Multivariate Geostatistics, 3rd edn. Springer, ch. 24-25.

Schabenberger, O. and Gotway, C. A. (2005). Statistical Methods for Spatial Data Analysis. Chapman and Hall/CRC.

Examples

set.seed(1)
cokrg(x = rnorm(50), y = rnorm(50), coords = matrix(runif(100), 50, 2), target = rnorm(50))
#> $estimate
#>  [1] -0.127268480  0.040549848 -0.054663291  0.211623892 -0.059660708
#>  [6]  0.380550364  0.107110214 -0.005168215  0.048962124 -0.809041377
#> [11]  0.258352997  0.117032319 -0.421964755  0.188766100  0.532085813
#> [16] -0.178288568  0.092213480  0.173977723  0.200650136  0.012699968
#> [21] -0.501731954 -0.293925544 -0.055739973  0.162143268  0.354335906
#> 
#> $se
#>  [1] 0.3193037 0.9379827 0.9896028 0.8711115 0.9861816 0.6072968 0.4156878
#>  [8] 0.9743784 0.9738287 0.5462035 0.9620072 0.4221227 0.7536308 0.9316495
#> [15] 0.6246962 0.9611224 0.9702036 0.5367493 0.9392200 0.9443456 0.5995348
#> [22] 0.8901429 0.9834200 0.9585632 0.7152172
#> 
#> $n
#> [1] 50
#> 
#> $method
#> [1] "Simple cokriging (exponential direct and cross covariances)"
#>