generate_gaussian_copula_samples: Generate Gaussian Copula Samples

In covalchemy: Constructing Joint Distributions with Control Over Statistical Properties

Generate Gaussian Copula Samples

Description

This function generates samples from a Gaussian copula given a specified correlation matrix. The samples are uniformly distributed in [0, 1] across dimensions.

Usage

generate_gaussian_copula_samples(n, d, rho_matrix)

Arguments

n	Integer. The number of samples to generate.
d	Integer. The dimensionality of the copula.
rho_matrix	A d x d positive-definite correlation matrix.

Details

The function works as follows:

1. Generates multivariate normal samples with the given correlation matrix.

2. Transforms the samples to the uniform distribution [0, 1] using the cumulative distribution function (CDF) of the standard normal.

Value

A matrix of size n x d, where each row represents a sample and each column corresponds to a dimension. The values are uniformly distributed in [0, 1].

Examples

# Example usage:
library(MASS)  # Load package for `mvrnorm`
rho_matrix <- matrix(c(1, 0.5, 0.5, 1), nrow = 2)  # 2x2 correlation matrix
samples <- generate_gaussian_copula_samples(n = 1000, d = 2, rho_matrix = rho_matrix)
head(samples)

covalchemy documentation built on April 12, 2025, 2:15 a.m.