Skip to contents

Multivariate Unimodality Index

boring.Rd

boring estimates how boring (i.e., unimodal) an empirical distribution is.

Usage

boring(x, w = rep(1, nrow(x)), na_rm = FALSE, conf_level = NA)

Arguments

x

A matrix with \(m\) columns and \(n\) rows, where each column represents a different variable and each row a different observation.

w

A optional non-negative and non-zero vector of weights for each observation. Its length must equal the number of rows of x.

na_rm

A logical indicating whether NA values should be stripped before the computation proceeds (default: FALSE).

conf_level

A scalar indicating the confidence level of the required confidence interval (default: NA, no confidence interval is returned). The confidence intervals are calculated via z-Transformation.

Value

A numeric value indicating how boring (i.e., unimodal) the empirical distribution is. Values close to 1 indicate that the distribution is likely unimodal.

Author

Simon Garnier, garnier@njit.edu

Examples

m1 <- matrix(c(rnorm(500, 6), rnorm(500, 11, 3)), ncol = 2)
m2 <- matrix(c(rnorm(500, 2), rnorm(500, 1, 1)), ncol = 2)
m3 <- matrix(c(rnorm(500, -13), rnorm(500, -3, 2)), ncol = 2)
X <- rbind(m1, m3)
boring(X)
#>   estimate 
#> -0.4399872