Weighted Variance

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

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)
#> [1] 9.166667
weighted_var(1:10, w = NULL)
#> [1] 9.166667
weighted_var(1:10, w = rep(1, 10))
#> [1] 9.166667
weighted_var(1:10, w = 1:10)
#> [1] 6.875
weighted_var(1:10, w = 1:10, method = "ML")
#> [1] 6