create_marginal_data_gaussian: Generate marginal Gaussian data using Cholesky decomposition

In shapr: Prediction Explanation with Dependence-Aware Shapley Values

Generate marginal Gaussian data using Cholesky decomposition

Description

Given a multivariate Gaussian distribution, this function creates data from specified marginals of said distribution.

Usage

create_marginal_data_gaussian(n_MC_samples, Sbar_features, mu, cov_mat)

Arguments

n_MC_samples	Integer. The number of samples to generate.
Sbar_features	Vector of integers indicating which marginals to sample from.
mu	Numeric vector containing the expected values for all features in the multivariate Gaussian distribution.
cov_mat	Numeric matrix containing the covariance between all features in the multivariate Gaussian distribution.

Author(s)

Lars Henry Berge Olsen

shapr documentation built on April 4, 2025, 12:18 a.m.