getCov: Get full covariance/correlation matrix from eclairs

In decorrelate: Decorrelation Projection Scalable to High Dimensional Data

Get full covariance/correlation matrix from eclairs

Description

Get full covariance/correlation matrix from eclairs decomposition

Usage

getCov(ecl, lambda, ...)

getCor(ecl, lambda, ...)

## S4 method for signature 'eclairs'
getCov(ecl, lambda, ...)

## S4 method for signature 'eclairs'
getCor(ecl, lambda, ...)

Arguments

ecl	eclairs decomposition
lambda	shrinkage parameter for the convex combination.
...	other arguments

Details

The full matrix is computationally expensive to compute and uses a lot of memory for large p. So it is better to use decorrelate or mult_eclairs to perform projections in O(np) time.

Value

p x p covariance/correlation matrix

Examples

library(Rfast)

n <- 800 # number of samples
p <- 200 # number of features

# create correlation matrix
Sigma <- autocorr.mat(p, .9)

# draw data from correlation matrix Sigma
Y <- rmvnorm(n, rep(0, p), sigma = Sigma * 5.1, seed = 1)
rownames(Y) <- paste0("sample_", seq(n))
colnames(Y) <- paste0("gene_", seq(p))

# eclairs decomposition
ecl <- eclairs(Y)

# extract covariance implied by eclairs decomposition
getCov(ecl)[1:3, 1:3]

decorrelate documentation built on Aug. 8, 2025, 7:55 p.m.