Description Usage Arguments Details Covariance functions See Also
Representation of one structure in a non stationary spatial continuity model.
1 2 3 4 | covarianceStructure3DNonStat(type = c("gaussian", "exponential", "spherical",
"cubic", "matern1", "matern2", "cauchy"), contribution, maxrange,
midrange = maxrange, minrange = midrange, azimuth = 0, dip = 0,
rake = 0, power = 1, control_points)
|
type |
The type of covariance function represented (see below). |
contribution |
The variance associated with this structure. |
maxrange, midrange, minrange |
The structure's range in three orthogonal directions, or the three semi-axes of the covariance ellipsoid. |
azimuth, dip, rake |
Orientation of the covariance ellipsoid in geological coordinates. |
power |
The power parameter for the covariance functions that feature one. |
control_points |
A |
All parameters except power
are non stationary, each
modeled by its own (stationary, noiseless) GP
and using the control points as
data. The GPs are initialized with the values provided as data, and by
default predict a constant value everywhere. Use the Fit
method of the
main GP to determine the best value for each parameter at the control points
and the stationary covariance model for each.
The following functions are available for the type
parameter. In the
equations below, d is the anisotropy-corrected distance between two points
and p is the power
parameter. Functions marked with an asterisk
support modeling with directional data.
gaussian
*exp(-3*d^2)
exponential
exp(-3*d)
spherical
1 - 1.5*d + 0.5*d^3, 0 <= d <= 1
cubic
*1 - 7*d^2 + 35/4*d^3 - 7/2*d^5 + 3/4*d^7, 0 <= d <= 1
matern1
*(1 + 5*d) * exp(-5*d)
matern2
*(1 + 6*d + 12*d^2) * exp(-6*d)
cauchy
*(1 + d^2)^(-p), p > 0
covarianceStructure3DNonStat-class
,
covarianceModel3D-class
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.