generate.patchwork.copula: Generate data from a non-functional dependence

In knnIndep: Independence tests and benchmarks

Description

Generate data from a non-functional dependence called 'patchwork copula'. Like a copula the data is uniform in x and y but it has a dependence between x and yy that has a block like structure

Usage

generate.patchwork.copula(p = matrix(rbeta(bins * bins, alpha, beta), ncol = bins),
	 alpha = 0.01, beta = 1, c = 1, npoints = 320, bins = 20, returnmi = FALSE, 
	 plot = FALSE)

Arguments

p	matrix, starting mass distribution on the grid
alpha	decimal, parameter of beta distribution used for p (if p left as per default)
beta	decimal, parameter of beta distribution used for p (if p left as per default)
c	decimal, concentration factor (default 1), used to stabilize mutual information estimation
npoints	decimal, sample size
bins	decimal, number of bins of the bins*bins grid
returnmi	boolean, whether to return the mutual information
plot	boolean, whether to plot the dependence

Value

list with the following elements

x	matrix of x-coordinates, each column corresponds to a noise level from noises
y	matrix of y-coordinates, each column corresponds to a noise level from noises
mi	mutual information of the dependence, only return if returnmi is set to TRUE

Author(s)

Sebastian Dümcke [e-mail]

Examples

generate.patchwork.copula(bins=20,plot=TRUE)

knnIndep documentation built on May 2, 2019, 3:23 a.m.