direction: Direction of a distribution

In extras: Helper Functions for Bayesian Analyses

Direction of a distribution

Description

The direction of a distribution is the side (left/right) that the distribution's center falls on, relative to a threshold. The center can be calculated using a user-specified function, including the median (default), mean, geometric mean, mode, or any other custom function. By convention, values below the threshold fall to the left, while values above the threshold fall to the right. Center estimates equal to the threshold are assumed to fall to the right.

Usage

direction(x, estimate = xtr_median, threshold = 0, na_rm = FALSE)

Arguments

x	A numeric vector of MCMC values or any other numeric vector of samples.
estimate	A function for estimating the center of the distribution. Defaults to xtr_median(), but can also be xtr_mean() or any custom function that returns a number (a non-missing numeric vector of length 1). If na_rm is true, NA values are dropped before calling the function.
threshold	A number of the threshold value.
na_rm	A flag specifying whether to remove missing values.

Value

A string indicating if at least half of the observations are above the threshold ("right") or not ("left"), or NA_character_ in the case of missing values when na_rm = FALSE.

See Also

Other summary: directional_information(), kurtosis(), lower(), probability_direction(), pvalue(), pzeros(), skewness(), svalue(), upper(), variance(), xtr_mean(), xtr_median(), xtr_sd(), zeros(), zscore()

Examples

direction(c(1, 2, 3))
direction(c(-1))
direction(c(0, 0, 0))
direction(c(-100, 1, 1))
direction(c(-100, 1, 1), mean)
direction(c(100, 0.01, 0.01), function(.x) exp(mean(log(.x))))

extras documentation built on July 16, 2026, 1:07 a.m.