Quick multivariate normal samples over two dimensions (cholesky precomputed)

This is equivalent to estimating the full spatio-temporal covariance matrix and sampling from the multivariate normal distribution: full_k <- kronecker(dist_k, time_k) f <- mvrnorm(1, rep(0, n * nt), full_k)

Usage

quick_mvnorm_chol(space_chol, time_chol)

Arguments

space_chol

Cholesky decomposition of sapace kernel matrix

time_chol

Cholesky decomposition of time kernel matrix