weighted_var: Weighted Variance

In MetricsWeighted: Weighted Metrics and Performance Measures for Machine Learning

Weighted Variance

Description

Calculates weighted variance, see stats::cov.wt() or https://en.wikipedia.org/wiki/Sample_mean_and_covariance#Weighted_samples for details.

Usage

weighted_var(x, w = NULL, method = c("unbiased", "ML"), na.rm = FALSE, ...)

Arguments

x	Numeric vector.
w	Optional vector of non-negative case weights.
method	Specifies how the result is scaled. If "unbiased", the denomiator is reduced by 1, see stats::cov.wt() for details.
na.rm	Should missing values in x be removed? Default is FALSE.
...	Further arguments passed to stats::cov.wt().

Value

A length-one numeric vector.

See Also

stats::cov.wt()

Examples

weighted_var(1:10)
weighted_var(1:10, w = NULL)
weighted_var(1:10, w = rep(1, 10))
weighted_var(1:10, w = 1:10)
weighted_var(1:10, w = 1:10, method = "ML")

MetricsWeighted documentation built on Nov. 16, 2023, 5:09 p.m.