data.simulation: Simulates subspace clustering data

In psobczyk/varclust: Variables Clustering

Description

Generates data for simulation with a low-rank subspace structure: variables are clustered and each cluster has a low-rank representation. Factors than span subspaces are not shared between clusters.

Usage

data.simulation(
  n = 100,
  SNR = 1,
  K = 10,
  numb.vars = 30,
  max.dim = 2,
  min.dim = 1,
  equal.dims = TRUE
)

Arguments

n	An integer, number of individuals.
SNR	A numeric, signal to noise ratio measured as variance of the variable, element of a subspace, to the variance of noise.
K	An integer, number of subspaces.
numb.vars	An integer, number of variables in each subspace.
max.dim	An integer, if equal.dims is TRUE then max.dim is dimension of each subspace. If equal.dims is FALSE then subspaces dimensions are drawn from uniform distribution on [min.dim,max.dim].
min.dim	An integer, minimal dimension of subspace .
equal.dims	A boolean, if TRUE (value set by default) all clusters are of the same dimension.

Value

A list consisting of:

X	matrix, generated data
signals	matrix, data without noise
dims	vector, dimensions of subspaces
factors	matrix, columns of which span subspaces
s	vector, true partiton of variables

Examples

sim.data <- data.simulation()
sim.data2 <- data.simulation(
  n = 30, SNR = 2, K = 5, numb.vars = 20,
  max.dim = 3, equal.dims = FALSE
)

psobczyk/varclust documentation built on June 18, 2021, 3:02 p.m.