Skip to contents

Wrapper for `lori::covmat`

Source: R/covmat.R
covmat.Rd

Thin helper that calls lori::covmat() without requiring users to attach the lori package. It constructs the block-structured covariate matrix \(X\) (with \(n \times p\) rows) from site-level covariates R, year-level covariates C, and optional site–year covariates E. Any of R, C, or E can be NULL to omit that block.

Usage

covmat(n, p, R = NULL, C = NULL, E = NULL, center = FALSE)

Arguments

n

Integer; number of sites.

p

Integer; number of years.

R

Optional numeric matrix with \(n\) rows and \(d_R\) columns (site-level covariates). Default: NULL.

C

Optional numeric matrix with \(p\) rows and \(d_C\) columns (year-level covariates). Default: NULL.

E

Optional numeric matrix with \(n p\) rows and \(d_E\) columns (site–year covariates; one row per site–year observation in the same order as \(X\)). Default: NULL.

center

Logical; whether to center columns within each block (as in lori::covmat). Default: FALSE.

Value

A numeric matrix with \(n \times p\) rows and \(d = d_R + d_C + d_E\) columns (when present), identical to the value returned by lori::covmat().

See also

covmat

Examples

if (requireNamespace("lori", quietly = TRUE)) {
  set.seed(1)
  n <- 3; p <- 2
  R <- matrix(rnorm(n * 2), nrow = n, ncol = 2)  # site-level (d_R = 2)
  C <- matrix(rnorm(p * 2), nrow = p, ncol = 2)  # year-level (d_C = 2)
  X <- covmat(n, p, R = R, C = C)
  dim(X)  # 6 x 4
}
#> [1] 6 4