cor_stat_rs: Calculate general stationary correlation.
In mcgf: Markov Chain Gaussian Fields Simulation and Parameter Estimation

cor_stat_rs

R Documentation

Calculate general stationary correlation.

Description

Calculate general stationary correlation.

Usage

cor_stat_rs(
  n_regime,
  base_ls,
  lagrangian_ls,
  par_base_ls,
  par_lagr_ls,
  lambda_ls,
  h_ls,
  h1_ls,
  h2_ls,
  u_ls,
  base_fixed = FALSE
)

Arguments

`n_regime`	Integer, number of regimes.
`base_ls`	List of base model, `sep` or `fs` for now. Or list of correlation matrices/arrays.
`lagrangian_ls`	List of Lagrangian model, `lagr_tri` or `lagr_askey` for now.
`par_base_ls`	List of parameters for the base model, used only when `base_fixed = FALSE`.
`par_lagr_ls`	List of parameters for the Lagrangian model. Used only when `lagrangian_ls` is not `none`.
`lambda_ls`	List of weight of the Lagrangian term, `\lambda\in[0, 1]`.
`h_ls`	List of Euclidean distance matrix or array, used only when `base_fixed = FALSE`.
`h1_ls`	List of horizontal distance matrix or array, same dimension as `h_ls`. Used only when `lagrangian_ls` is not `none`.
`h2_ls`	List of vertical distance matrix or array, same dimension as `h_ls`. Used only when `lagrangian_ls` is not `none`.
`u_ls`	List of time lag, same dimension as `h_ls`.
`base_fixed`	Logical; if TRUE, `base_ls` is the list of correlation.

Details

It gives a list of general stationary correlation for n_regime regimes. See cor_stat for the model details. Model parameters are lists of length 1 or n_regime. When length is 1, same values are used for all regimes. If base_fixed = TRUE, the base is a list of correlation and par_base_ls and h_ls are not used.

Value

Correlations for the general stationary model. Same dimension of base_ls if base_fixed = TRUE.

Examples

# Fit general stationary model with sep base.
par_s <- list(nugget = 0.5, c = 0.01, gamma = 0.5)
par_t <- list(a = 1, alpha = 0.5)
par_base <- list(par_s = par_s, par_t = par_t)
h1 <- array(c(0, 5, -5, 0), dim = c(2, 2, 3))
h2 <- array(c(0, 8, -8, 0), dim = c(2, 2, 3))
h <- sqrt(h1^2 + h2^2)
u <- array(rep(c(1, 2, 3), each = 4), dim = c(2, 2, 3))
cor_stat_rs(
    n_regime = 2,
    base_ls = list("sep"),
    lagrangian_ls = list("none", "lagr_tri"),
    par_base_ls = list(par_base),
    par_lagr_ls = list(NULL, list(v1 = 10, v2 = 20)),
    lambda_ls = list(0, 0.2),
    h_ls = list(h),
    h1_ls = list(NULL, h1),
    h2_ls = list(NULL, h2),
    u_ls = list(u, u + 1)
)

# Fit general stationary model given fs as the base model.
h1 <- array(c(0, 5, -5, 0), dim = c(2, 2, 3))
h2 <- array(c(0, 8, -8, 0), dim = c(2, 2, 3))
h <- sqrt(h1^2 + h2^2)
u <- array(rep(c(0.1, 0.2, 0.3), each = 4), dim = c(2, 2, 3))
fit_base <- cor_fs(
    nugget = 0.5, c = 0.01, gamma = 0.5, a = 1, alpha = 0.5,
    beta = 0.0, h = h, u = u
)
par_lagr <- list(v1 = 5, v2 = 10)
cor_stat_rs(
    n_regime = 2,
    par_lagr_ls = list(par_lagr),
    h1_ls = list(h1),
    h2_ls = list(h2),
    u_ls = list(u, u + 1),
    lambda_ls = list(0, 0.8),
    base_ls = list(fit_base),
    lagrangian = list("lagr_tri", "lagr_askey"),
    base_fixed = TRUE
)

mcgf documentation built on June 29, 2024, 9:09 a.m.