scatter: Compute the scatter matrix

In AEBilgrau/correlateR: Fast correlations and covariances

Description

This functions is a fast computation of the scatter matrix and cross-scatter matrix given by

XX^T

and XY^T respectively.

Usage

scatter(X, center = TRUE)

xscatter(X, Y, center = TRUE)

Arguments

X	An n by p1 numeric matrix with observations in the rows and variables in the columns.
center	logical. If TRUE, the data-matrices are centered first.
Y	An n by p2 as X.

Value

The p1 times p1 scatter matrix or p1 times p2 cross-scatter matrix.

Author(s)

Anders Ellern Bilgrau <anders.ellern.bilgrau (at) gmail.com>

Examples

n <- 11
X <- createData(n, 4)
Y <- createData(n, 2)
scatter(X)
(n-1)*cov(X)

xscatter(X, Y)
n*xcov(X, Y, method = "ML")  # == (n - 1)*xcov(X, Y)

AEBilgrau/correlateR documentation built on Nov. 15, 2019, 9:21 a.m.