latin.hypercube: Latin hypercube design matrix

In emulator: Bayesian Emulation of Computer Programs

Latin hypercube design matrix

Description

Gives a Latin hypercube design matrix with an arbitrary number of points in an arbitrary number of dimensions. The toy dataset was generated using latin.hypercube().

Usage

latin.hypercube(n, d, names=NULL, normalize=FALSE, complex=FALSE)

Arguments

n	Number of points
d	Number of dimensions
names	Character vector for column names (optional)
normalize	Boolean variable with TRUE meaning to normalize each column so the minimum is zero and the maximum is one. If it takes its default FALSE, the points represent midpoints of n equispaced intervals; the points thus have a minimum of 0.5/n and a maximum of 1-0.5/n
complex	Boolean with default FALSE meaning to return a complex latin hypercube in which real and imaginary components separately form a latin hypercube

Author(s)

Robin K. S. Hankin

Examples

#10 points, 6 dimensions:
(latin.hypercube(10,6) -> x)
plot(as.data.frame(x))

latin.hypercube(10,2,complex=TRUE)

emulator documentation built on May 29, 2024, 6:18 a.m.