generate_gb_covariance: Generate a Graph-Banded Covariance Matrix

In jacobbien/ggb: Graph-Guided Banding of the Covariance Matrix

Description

Returns a p-by-p matrix covariance matrix that is banded with respect to the graph g and with local bandwidths given in the p-vector b.

Usage

generate_gb_covariance(g, b, sigma = 0.01, cor = TRUE)

Arguments

g	a seed graph (with p vertices)
b	a p-vector of local bandwidths
sigma	sd of independent error to add to each node
cor	whether to scale rows/cols to have variance 1

Examples

g <- igraph::graph.lattice(c(5, 4))
b <- rep(1, 20)
Sig <- generate_gb_covariance(g, b)

jacobbien/ggb documentation built on May 18, 2019, 8:01 a.m.