covcor_matrix: Multivariate Gaussian covariance/correlation matrices

In jbrowell/ProbCast: Flexible Probabilistic Forecasting

Multivariate Gaussian covariance/correlation matrices

Description

This function produces a list of Multivariate Gaussian spatial/tempral/spatiotemporal covarance/correlation matrices from uniform distributed variables.

Usage

covcor_matrix(
  u_data,
  kfold = NULL,
  cov_cor = "covariance",
  use = "pairwise.complete.obs",
  boundary_threshold = NA,
  forcePD = F,
  scale = F,
  ...
)

Arguments

u_data	A dataframe of standard uniform distributed variables.
kfold	A vector of kfold identifiers.
cov_cor	specify either covariance or correlation.
boundary_threshold	Data outside [0,1] are set to boundary_threshold or 1-boundary_threshold. Set boundary_threshold=NA to exclude data outside [0,1].
forcePD	Use nearPD() to return a positive definite cov/cor matrix
scale	Scale gaussian variable inputs to have unit variance. for covariance matrix? Defaults to FALSE.

Value

A list of covariance/correlation matrices corresponding to unique values of kfold

Author(s)

Jethro Browell, [e-mail]; Ciaran Gilbert, [e-mail]

jbrowell/ProbCast documentation built on July 20, 2024, 1:53 p.m.