Skip to contents

Geometric centroid of n objects

Source: R/divergences_generic.R
geomcent.Rd

geomcent() is an S3 generic for computing the geometric mean of R objects, with methods for the following classes:

  • igcmatcatdist

Usage

geomcent(p, weight = NULL, base = exp(1))

Arguments

p

An R object gathering a number n of objects of same nature. Currently, there is a method for igcmatcatdist matrix, i.e., a matrix of categorical distribution.

weight

A positive vector of length n. The geometric mean is balanced by weighting each row of p with the corresponding element of weight.

base

A positive value specifying the base of logarithm used for computing Jeffreys divergence. Default is exp(1) (natural logarithm).

Value

The geometric centroid of the input objects p weighted by weight.

See also

arithcent(), jcent(), jscent(), catdist.

Examples

p <- igccatdist(dbinom(0:4, size = 4, prob = 0.5))
q <- igccatdist(rep(1/5, times = 5))
matp <- matrix(c(p, q), nrow = 2)
matptest <- igcmatcatdist(p = matp)
geomcent(p = matptest)
#> [1] 0.1250000 0.3061862 0.1118034 0.2000000 0.2000000
#> attr(,"class")
#> [1] "igccatdist"
#> attr(,"mass")
#> [1] 0.9429896