.cor_stat | R Documentation |
Calculate general stationary correlation.
.cor_stat(base, lagrangian, par_base, par_lagr, lambda, base_fixed = FALSE)
base |
Base model, |
lagrangian |
Lagrangian model, |
par_base |
Parameters for the base model (symmetric), used only when
|
par_lagr |
Parameters for the Lagrangian model. Used only when
|
lambda |
Weight of the Lagrangian term, |
base_fixed |
Logical; if TRUE, |
The general station model, a convex combination of a base model and a Lagrangian model, has the form
C(\mathbf{h}, u)=(1-\lambda)C_{\text{Base}}(\mathbf{h}, u)+
\lambda C_{\text{Lagr}}(\mathbf{h}, u),
where \lambda
is the weight of the Lagrangian term.
If base_fixed = TRUE
, the correlation is of the form
C(\mathbf{h}, u)=(1-\lambda)C_{\text{Base}}+
\lambda C_{\text{Lagr}}(\mathbf{h}, u),
where base
is a correlation matrix/array and par_base
and h
are not
used.
When lagrangian = "none"
, lambda
must be 0.
Correlations for the general stationary model. Same dimension of
base
if base_fixed = FALSE
.
Gneiting, T., Genton, M., & Guttorp, P. (2006). Geostatistical Space-Time Models, Stationarity, Separability, and Full Symmetry. In C&H/CRC Monographs on Statistics & Applied Probability (pp. 151–175). Chapman and Hall/CRC.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.