smtools: Some Tools for Handling Symmetric Matrices

In corpcor: Efficient Estimation of Covariance and (Partial) Correlation

Description

sm2vec takes a symmetric matrix and puts the lower triagonal entries into a vector (cf. lower.tri).

sm.index lists the corresponding x-y-indices for each entry in the vector produced by sm2vec.

vec2sm reverses the operation by sm2vec and converts the vector back to a symmetric matrix. If diag=FALSE the diagonal of the resulting matrix will consist of NAs. If order is supplied then the input vector vec will first be rearranged accordingly.

Usage

sm2vec(m, diag = FALSE)
sm.index(m, diag = FALSE)
vec2sm(vec, diag = FALSE, order = NULL)

Arguments

m	symmetric matrix
diag	logical. Should the diagonal be included in the conversion to and from a vector?
vec	vector of unique elements from a symmetric matrix
order	order of the entries in vec

Value

A vector (sm2vec), a two-column matrix with indices (sm.index), or a symmetric matrix (vec2sm).

Author(s)

Korbinian Strimmer (https://strimmerlab.github.io/).

See Also

lower.tri.

Examples

# load corpcor library
library("corpcor")

# a symmetric matrix
m = rbind(
 c(3,1,1,0),
 c(1,3,0,1),
 c(1,0,2,0),
 c(0,1,0,2)
)
m

# convert into vector (including the diagonals)
v = sm2vec(m, diag=TRUE)
v.idx = sm.index(m, diag=TRUE)
v
v.idx

# put back to symmetric matrix
vec2sm(v, diag=TRUE)

# convert from vector with specified order of the elements
sv = sort(v)
sv
ov = order(v)
ov
vec2sm(sv, diag=TRUE, order=ov)

corpcor documentation built on Sept. 16, 2021, 5:12 p.m.